Search Results for author:
Found 329 papers, 133 papers with code
Unconstrained Realtime Facial Performance Capture
We introduce a realtime facial tracking system specifically designed for performance capture in unconstrained settings using a consumer-level RGB-D sensor.
Dense Human Body Correspondences Using Convolutional Networks
We propose a deep learning approach for finding dense correspondences between 3D scans of people.
Real-Time Facial Segmentation and Performance Capture from RGB Input
We adopt a state-of-the-art regression-based facial tracking framework with segmented face images as training, and demonstrate accurate and uninterrupted facial performance capture in the presence of extreme occlusion and even side views.
Capturing Dynamic Textured Surfaces of Moving Targets
We present an end-to-end system for reconstructing complete watertight and textured models of moving subjects such as clothed humans and animals, using only three or four handheld sensors.
Pruning Filters for Efficient ConvNets
However, magnitude-based pruning of weights reduces a significant number of parameters from the fully connected layers and may not adequately reduce the computation costs in the convolutional layers due to irregular sparsity in the pruned networks.
Ranked #1 on
Network Pruning
on ImageNet
Deep CTR Prediction in Display Advertising
Click through rate (CTR) prediction of image ads is the core task of online display advertising systems, and logistic regression (LR) has been frequently applied as the prediction model.
Production-Level Facial Performance Capture Using Deep Convolutional Neural Networks
We present a real-time deep learning framework for video-based facial performance capture -- the dense 3D tracking of an actor's face given a monocular video.
High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis
Recent advances in deep learning have shown exciting promise in filling large holes in natural images with semantically plausible and context aware details, impacting fundamental image manipulation tasks such as object removal.
Photorealistic Facial Texture Inference Using Deep Neural Networks
We present a data-driven inference method that can synthesize a photorealistic texture map of a complete 3D face model given a partial 2D view of a person in the wild.
3D Face Reconstruction with Geometry Details from a Single Image
3D face reconstruction from a single image is a classical and challenging problem, with wide applications in many areas.
On the Effects of Batch and Weight Normalization in Generative Adversarial Networks
We further demonstrate the stability of WN on a 21-layer ResNet trained with the CelebA data set.
Training Quantized Nets: A Deeper Understanding
Currently, deep neural networks are deployed on low-power portable devices by first training a full-precision model using powerful hardware, and then deriving a corresponding low-precision model for efficient inference on such systems.
Adaptive Consensus ADMM for Distributed Optimization
The alternating direction method of multipliers (ADMM) is commonly used for distributed model fitting problems, but its performance and reliability depend strongly on user-defined penalty parameters.
Auto-Conditioned Recurrent Networks for Extended Complex Human Motion Synthesis
We present a real-time method for synthesizing highly complex human motions using a novel training regime we call the auto-conditioned Recurrent Neural Network (acRNN).
Inspiring Computer Vision System Solutions
Reviewing this project with modern eyes provides us with the opportunity to reflect on several issues, relevant now as then to the field of computer vision and research in general, that go beyond the technical aspects of the work.
Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMM
Although deep learning models are highly effective for various learning tasks, their high computational costs prohibit the deployment to scenarios where either memory or computational resources are limited.
MEMEN: Multi-layer Embedding with Memory Networks for Machine Comprehension
Machine comprehension(MC) style question answering is a representative problem in natural language processing.
Ranked #37 on
Question Answering
on TriviaQA
Learning Dense Facial Correspondences in Unconstrained Images
To train such a network, we generate a massive dataset of synthetic faces with dense labels using renderings of a morphable face model with variations in pose, expressions, lighting, and occlusions.
Using Optimal Ratio Mask as Training Target for Supervised Speech Separation
In this paper, we use the optimal ratio mask as the training target of the deep neural network (DNN) for speech separation.
Realistic Dynamic Facial Textures From a Single Image Using GANs
By retargeting the PCA expression geometry from the source, as well as using the newly inferred texture, we can both animate the face and perform video face replacement on the source video using the target appearance.
Performance Prediction and Optimization of Solar Water Heater via a Knowledge-Based Machine Learning Method
This Chapter consists of: i) Comparative studies on varieties of machine learning models (artificial neural networks (ANNs), support vector machine (SVM) and extreme learning machine (ELM)) to predict the performances of SWHs; ii) Development of an ANN-based software to assist the quick prediction and iii) Introduction of a computational HTS method to design a high-performance SWH system.
Learning a model of facial shape and expression from 4D scans
FLAME is low-dimensional but more expressive than the FaceWarehouse model and the Basel Face Model.
Ranked #3 on
Face Alignment
on FaceScape
Keyword-based Query Comprehending via Multiple Optimized-Demand Augmentation
In this paper, we propose a novel neural network system that consists a Demand Optimization Model based on a passage-attention neural machine translation and a Reader Model that can find the answer given the optimized question.
Robust and Precise Vehicle Localization based on Multi-sensor Fusion in Diverse City Scenes
We present a robust and precise localization system that achieves centimeter-level localization accuracy in disparate city scenes.
Contextual-based Image Inpainting: Infer, Match, and Translate
We study the task of image inpainting, which is to fill in the missing region of an incomplete image with plausible contents.
Visualizing the Loss Landscape of Neural Nets
Neural network training relies on our ability to find "good" minimizers of highly non-convex loss functions.
Restricted Deformable Convolution based Road Scene Semantic Segmentation Using Surround View Cameras
Finally, an RDC based semantic segmentation model is built; the model is trained for real-world surround view images through a multi-task learning architecture by combining real-world images with transformed images.
On Modular Training of Neural Acoustics-to-Word Model for LVCSR
Finally, modules are integrated into an acousticsto-word model (A2W) and jointly optimized using acoustic data to retain the advantage of sequence modeling.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
DoubleFusion: Real-time Capture of Human Performances with Inner Body Shapes from a Single Depth Sensor
We further propose a joint motion tracking method based on the double layer representation to enable robust and fast motion tracking performance.
Robust Optimization over Multiple Domains
Hence, we propose to learn the model and the adversarial distribution simultaneously with the stochastic algorithm for efficiency.
Anime Style Space Exploration Using Metric Learning and Generative Adversarial Networks
Deep learning-based style transfer between images has recently become a popular area of research.
Large-scale Distance Metric Learning with Uncertainty
Furthermore, we can show that the metric is learned from latent examples only, but it can preserve the large margin property even for the original data.
Mesoscopic Facial Geometry Inference Using Deep Neural Networks
We present a learning-based approach for synthesizing facial geometry at medium and fine scales from diffusely-lit facial texture maps.
Learning with Structured Representations for Negation Scope Extraction
We report an empirical study on the task of negation scope extraction given the negation cue.
HybridFusion: Real-Time Performance Capture Using a Single Depth Sensor and Sparse IMUs
We propose a light-weight and highly robust real-time human performance capture method based on a single depth camera and sparse inertial measurement units (IMUs).
Deep Volumetric Video From Very Sparse Multi-View Performance Capture
We present a deep learning-based volumetric capture approach for performance capture using a passive and highly sparse multi-view capture system.
Real-Time Hair Rendering using Sequential Adversarial Networks
To handle the diversity of hairstyles and its appearance complexity, we disentangle hair structure, color, and illumination properties using a sequential GAN architecture and a semi-supervised training approach.
Spelling Error Correction Using a Nested RNN Model and Pseudo Training Data
We propose a nested recurrent neural network (nested RNN) model for English spelling error correction and generate pseudo data based on phonetic similarity to train it.
Multivariate Spatiotemporal Hawkes Processes and Network Reconstruction
There is often latent network structure in spatial and temporal data and the tools of network analysis can yield fascinating insights into such data.
Deep RBFNet: Point Cloud Feature Learning using Radial Basis Functions
Three-dimensional object recognition has recently achieved great progress thanks to the development of effective point cloud-based learning frameworks, such as PointNet and its extensions.
On the Continuity of Rotation Representations in Neural Networks
Thus, widely used representations such as quaternions and Euler angles are discontinuous and difficult for neural networks to learn.
SiCloPe: Silhouette-Based Clothed People
The synthesized silhouettes which are the most consistent with the input segmentation are fed into a deep visual hull algorithm for robust 3D shape prediction.
Soft Rasterizer: Differentiable Rendering for Unsupervised Single-View Mesh Reconstruction
We also show that our soft rasterizer can achieve comparable results to the cutting-edge supervised learning method and in various cases even better ones, especially for real-world data.
Pancreas segmentation with probabilistic map guided bi-directional recurrent UNet
To alleviate these problems, we propose a probabilistic-map-guided bi-directional recurrent UNet (PBR-UNet) architecture, which fuses intra-slice information and inter-slice probabilistic maps into a local 3D hybrid regularization scheme, which is followed by bi-directional recurrent network optimization.
A Model-Driven Stack-Based Fully Convolutional Network for Pancreas Segmentation
The irregular geometry and high inter-slice variability in computerized tomography (CT) scans of the human pancreas make an accurate segmentation of this crucial organ a challenging task for existing data-driven deep learning methods.
Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoning
Rendering bridges the gap between 2D vision and 3D scenes by simulating the physical process of image formation.
Ranked #1 on
3D Object Reconstruction
on ShapeNet
Transformable Bottleneck Networks
We propose a novel approach to performing fine-grained 3D manipulation of image content via a convolutional neural network, which we call the Transformable Bottleneck Network (TBN).
Semi-Supervised First-Person Activity Recognition in Body-Worn Video
Body-worn cameras are now commonly used for logging daily life, sports, and law enforcement activities, creating a large volume of archived footage.
PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization
We introduce Pixel-aligned Implicit Function (PIFu), a highly effective implicit representation that locally aligns pixels of 2D images with the global context of their corresponding 3D object.
Ranked #1 on
3D Object Reconstruction
on RenderPeople
Learning Perspective Undistortion of Portraits
In contrast to the previous state-of-the-art approach, our method handles even portraits with extreme perspective distortion, as we avoid the inaccurate and error-prone step of first fitting a 3D face model.
Disentangling Style and Content in Anime Illustrations
We propose a novel Generative Adversarial Disentanglement Network which can disentangle two complementary factors of variations when only one of them is labelled in general, and fully decompose complex anime illustrations into style and content in particular.
Neural Chinese Address Parsing
This paper introduces a new task {--} Chinese address parsing {--} the task of mapping Chinese addresses into semantically meaningful chunks.
Robust Gaussian Process Regression for Real-Time High Precision GPS Signal Enhancement
We validate the superiority of the proposed method in our real-time high precision positioning system against several popular state-of-the-art robust regression methods.
HPILN: A feature learning framework for cross-modality person re-identification
Most video surveillance systems use both RGB and infrared cameras, making it a vital technique to re-identify a person cross the RGB and infrared modalities.
Cross-Modality Person Re-identification
Person Re-Identification
MacNet: Transferring Knowledge from Machine Comprehension to Sequence-to-Sequence Models
In this paper, we transfer knowledge learned from machine comprehension to the sequence-to-sequence tasks to deepen the understanding of the text.
DR Loss: Improving Object Detection by Distributional Ranking
This architecture is efficient but can suffer from the imbalance issue with respect to two aspects: the inter-class imbalance between the number of candidates from foreground and background classes and the intra-class imbalance in the hardness of background candidates, where only a few candidates are hard to be identified.
X-LineNet: Detecting Aircraft in Remote Sensing Images by a pair of Intersecting Line Segments
Motivated by the development of deep convolution neural networks (DCNNs), tremendous progress has been gained in the field of aircraft detection.
Reinforced Dynamic Reasoning for Conversational Question Generation
This paper investigates a new task named Conversational Question Generation (CQG) which is to generate a question based on a passage and a conversation history (i. e., previous turns of question-answer pairs).
Improved Techniques for Training Adaptive Deep Networks
Adaptive inference is a promising technique to improve the computational efficiency of deep models at test time.
SoftTriple Loss: Deep Metric Learning Without Triplet Sampling
The set of triplet constraints has to be sampled within the mini-batch.
Ranked #21 on
Metric Learning
on CUB-200-2011
(using extra training data)
Learning Explicit and Implicit Structures for Targeted Sentiment Analysis
In this work, we argue that both types of information (implicit and explicit structural information) are crucial for building a successful targeted sentiment analysis model.
Learning to Infer Implicit Surfaces without 3D Supervision
The representation of 3D surfaces itself is a key factor for the quality and resolution of the 3D output.
Hierarchically Robust Representation Learning
Experiments on benchmark data sets demonstrate the effectiveness of the robust deep representations.
So2Sat LCZ42: A Benchmark Dataset for Global Local Climate Zones Classification
This is especially true for an automated analysis of remote sensing images on a global scale, which enables us to address global challenges such as urbanization and climate change using state-of-the-art machine learning techniques.
BIG-bench Machine Learning
Cultural Vocal Bursts Intensity Prediction
+1
Oriented Objects as pairs of Middle Lines
It is noteworthy that the objects in COCO can be regard as a special form of oriented objects with an angle of 90 degrees.
Ranked #13 on
Oriented Object Detection
on DOTA 1.0
Objects detection for remote sensing images based on polar coordinates
In this article, we introduce the polar coordinate system to the deep learning detector for the first time, and propose an anchor free Polar Remote Sensing Object Detector (P-RSDet), which can achieve competitive detection accuracy via uses simpler object representation model and less regression parameters.
Rethinking the Hyperparameters for Fine-tuning
Our findings challenge common practices of fine-tuning and encourages deep learning practitioners to rethink the hyperparameters for fine-tuning.
Learning Formation of Physically-Based Face Attributes
Based on a combined data set of 4000 high resolution facial scans, we introduce a non-linear morphable face model, capable of producing multifarious face geometry of pore-level resolution, coupled with material attributes for use in physically-based rendering.
Attribute Mix: Semantic Data Augmentation for Fine Grained Recognition
In this paper, we propose Attribute Mix, a data augmentation strategy at attribute level to expand the fine-grained samples.
Ranked #22 on
Fine-Grained Image Classification
on CUB-200-2011
CSRN: Collaborative Sequential Recommendation Networks for News Retrieval
Recurrent Neural Network (RNN)-based sequential recommendation is a popular approach that utilizes users' recent browsing history to predict future items.
ARCH: Animatable Reconstruction of Clothed Humans
In this paper, we propose ARCH (Animatable Reconstruction of Clothed Humans), a novel end-to-end framework for accurate reconstruction of animation-ready 3D clothed humans from a monocular image.
Ranked #3 on
3D Object Reconstruction From A Single Image
on BUFF
3D Object Reconstruction From A Single Image
3D Reconstruction
Intuitive, Interactive Beard and Hair Synthesis with Generative Models
We present an interactive approach to synthesizing realistic variations in facial hair in images, ranging from subtle edits to existing hair to the addition of complex and challenging hair in images of clean-shaven subjects.
Lightweight Mask R-CNN for Long-Range Wireless Power Transfer Systems
The proposed approach makes the object detection much easier to be transplanted on mobile devices and reduce the burden of hardware computation.
Multi-Domain Learning and Identity Mining for Vehicle Re-Identification
Our solution is based on a strong baseline with bag of tricks (BoT-BS) proposed in person ReID.
One-Shot Identity-Preserving Portrait Reenactment
This is achieved by a novel landmark disentanglement network (LD-Net), which predicts personalized facial landmarks that combine the identity of the target with expressions and poses from a different subject.
NTIRE 2020 Challenge on Real-World Image Super-Resolution: Methods and Results
This paper reviews the NTIRE 2020 challenge on real world super-resolution.
Generative Tweening: Long-term Inbetweening of 3D Human Motions
We introduce a biomechanically constrained generative adversarial network that performs long-term inbetweening of human motions, conditioned on keyframe constraints.
Weakly Supervised Representation Learning with Coarse Labels
To mitigate this challenge, we propose an algorithm to learn the fine-grained patterns for the target task, when only its coarse-class labels are available.
Fully Convolutional Mesh Autoencoder using Efficient Spatially Varying Kernels
Learning latent representations of registered meshes is useful for many 3D tasks.
Why Attentions May Not Be Interpretable?
In this paper, we demonstrate that one root cause of this phenomenon is the combinatorial shortcuts, which means that, in addition to the highlighted parts, the attention weights themselves may carry extra information that could be utilized by downstream models after attention layers.
Revisiting the Continuity of Rotation Representations in Neural Networks
In this paper, we provide some careful analysis of certain pathological behavior of Euler angles and unit quaternions encountered in previous works related to rotation representation in neural networks.
Towards Understanding Label Smoothing
Label smoothing regularization (LSR) has a great success in training deep neural networks by stochastic algorithms such as stochastic gradient descent and its variants.
Neural Architecture Design for GPU-Efficient Networks
To address this issue, we propose a general principle for designing GPU-efficient networks based on extensive empirical studies.
Efficient Graph-Based Active Learning with Probit Likelihood via Gaussian Approximations
We present a novel adaptation of active learning to graph-based semi-supervised learning (SSL) under non-Gaussian Bayesian models.
Posterior Consistency of Semi-Supervised Regression on Graphs
Graph-based semi-supervised regression (SSR) is the problem of estimating the value of a function on a weighted graph from its values (labels) on a small subset of the vertices.
Monocular Real-Time Volumetric Performance Capture
We present the first approach to volumetric performance capture and novel-view rendering at real-time speed from monocular video, eliminating the need for expensive multi-view systems or cumbersome pre-acquisition of a personalized template model.
Learning to Generate Diverse Dance Motions with Transformer
We also introduce new evaluation metrics for the quality of synthesized dance motions, and demonstrate that our system can outperform state-of-the-art methods.
OptMatch: Optimized Matchmaking via Modeling the High-Order Interactions on the Arena
OptMatch contains an offline learning stage and an online planning stage.
Stochastic scattering theory for excitation induced dephasing: Comparison to the Anderson-Kubo lineshape
In this paper we present a quantum stochastic model for spectroscopic line-shapes in the presence of a co-evolving and non-stationary background population of excitations.
Chemical Physics Mesoscale and Nanoscale Physics
A Federated Multi-View Deep Learning Framework for Privacy-Preserving Recommendations
Federated Learning(FL) is a newly developed privacy-preserving machine learning paradigm to bridge data repositories without compromising data security and privacy.
Hybrid Differentially Private Federated Learning on Vertically Partitioned Data
We present HDP-VFL, the first hybrid differentially private (DP) framework for vertical federated learning (VFL) to demonstrate that it is possible to jointly learn a generalized linear model (GLM) from vertically partitioned data with only a negligible cost, w. r. t.
Semi-Anchored Detector for One-Stage Object Detection
The anchor-free strategy benefits the classification task but can lead to sup-optimum for the regression task due to the lack of prior bounding boxes.
AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results
This paper introduces the real image Super-Resolution (SR) challenge that was part of the Advances in Image Manipulation (AIM) workshop, held in conjunction with ECCV 2020.
Improved Knowledge Distillation via Full Kernel Matrix Transfer
Recently, a number of works propose to transfer the pairwise similarity between examples to distill relative information.
Dynamic Facial Asset and Rig Generation from a Single Scan
We also model the joint distribution between identities and expressions, enabling the inference of the full set of personalized blendshapes with dynamic appearances from a single neutral input scan.
WeMix: How to Better Utilize Data Augmentation
To this end, we develop two novel algorithms, termed "AugDrop" and "MixLoss", to correct the data bias in the data augmentation.
Position-Aware Tagging for Aspect Sentiment Triplet Extraction
Our observation is that the three elements within a triplet are highly related to each other, and this motivates us to build a joint model to extract such triplets using a sequence tagging approach.
Ranked #3 on
Aspect Sentiment Triplet Extraction
on SemEval
Domain Agnostic Learning for Unbiased Authentication
We further extend our method to a meta-learning framework to pursue more thorough domain-difference elimination.
Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation
In this paper, we propose to automate the design of metric-specific loss functions by searching differentiable surrogate losses for each metric.
Learning Accurate Entropy Model with Global Reference for Image Compression
In this work, we propose a novel Global Reference Model for image compression to effectively leverage both the local and the global context information, leading to an enhanced compression rate.
Speakerfilter-Pro: an improved target speaker extractor combines the time domain and frequency domain
This paper introduces an improved target speaker extractor, referred to as Speakerfilter-Pro, based on our previous Speakerfilter model.
Data Augmentation for End-to-end Code-switching Speech Recognition
Training a code-switching end-to-end automatic speech recognition (ASR) model normally requires a large amount of data, while code-switching data is often limited.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+4
Center-wise Local Image Mixture For Contrastive Representation Learning
Contrastive learning based on instance discrimination trains model to discriminate different transformations of the anchor sample from other samples, which does not consider the semantic similarity among samples.
ULSD: Unified Line Segment Detection across Pinhole, Fisheye, and Spherical Cameras
Targeting at the unified line segment detection (ULSD) for both distorted and undistorted images, we propose to represent line segments with the Bezier curve model.
Ranked #5 on
Line Segment Detection
on wireframe dataset
(sAP10 metric)
Heterogeneous Contrastive Learning: Encoding Spatial Information for Compact Visual Representations
Contrastive learning has achieved great success in self-supervised visual representation learning, but existing approaches mostly ignored spatial information which is often crucial for visual representation.
Arbitrary-Oriented Object Detection in Remote Sensing Images Based on Polar Coordinates
In this article, we introduce the polar coordinate system to the deep learning detector for the first time, and propose an anchor free Polar Remote Sensing Object Detector (P-RSDet), which can achieve competitive detection accuracy via using simpler object representation model and less regression parameters.
Ranked #12 on
Oriented Object Detection
on DOTA 1.0
Seed the Views: Hierarchical Semantic Alignment for Contrastive Representation Learning
In this paper, we propose a hierarchical semantic alignment strategy via expanding the views generated by a single image to \textbf{Cross-samples and Multi-level} representation, and models the invariance to semantically similar images in a hierarchical way.
An Event Correlation Filtering Method for Fake News Detection
To improve the detection performance of fake news, we take advantage of the event correlations of news and propose an event correlation filtering method (ECFM) for fake news detection, mainly consisting of the news characterizer, the pseudo label annotator, the event credibility updater, and the news entropy selector.
A Coarse-to-Fine Auto-Sampler For Long-tailed Image Recognition
The long-tail distributed data in the real world has always been a great challenge for deep learning.
AU-Guided Unsupervised Domain Adaptive Facial Expression Recognition
The domain diversities including inconsistent annotation and varied image collection conditions inevitably exist among different facial expression recognition (FER) datasets, which pose an evident challenge for adapting the FER model trained on one dataset to another one.
Facial Expression Recognition
Facial Expression Recognition (FER)
Blurring Fools the Network -- Adversarial Attacks by Feature Peak Suppression and Gaussian Blurring
Existing pixel-level adversarial attacks on neural networks may be deficient in real scenarios, since pixel-level changes on the data cannot be fully delivered to the neural network after camera capture and multiple image preprocessing steps.
Amplifying the Anterior-Posterior Difference via Data Enhancement -- A More Robust Deep Monocular Orientation Estimation Solution
Existing deep-learning based monocular orientation estimation algorithms faces the problem of confusion between the anterior and posterior parts of the objects, caused by the feature similarity of such parts in typical objects in traffic scenes such as cars and pedestrians.
Exploiting Vulnerability of Pooling in Convolutional Neural Networks by Strict Layer-Output Manipulation for Adversarial Attacks
Then an attack method based on Strict Pooling Manipulation (SPM) which is an instantiation of the SLOM spirit is designed to effectively realize both type I and type II adversarial attacks on a target CNN.
AsymptoticNG: A regularized natural gradient optimization algorithm with look-ahead strategy
Optimizers that further adjust the scale of gradient, such as Adam, Natural Gradient (NG), etc., despite widely concerned and used by the community, are often found poor generalization performance, compared with Stochastic Gradient Descent (SGD).
1st Place Solution to VisDA-2020: Bias Elimination for Domain Adaptive Pedestrian Re-identification
Considering the large gap between the source domain and target domain, we focused on solving two biases that influenced the performance on domain adaptive pedestrian Re-ID and proposed a two-stage training procedure.
Detection of magnetic gap in the topological surface states of MnBi2Te4
An essential ingredient to realize these quantum states is the magnetic gap in the topological surface states induced by the out-of-plane ferromagnetism on the surface of MnBi2Te4.
Materials Science
Zen-NAS: A Zero-Shot NAS for High-Performance Image Recognition
To address this issue, instead of using an accuracy predictor, we propose a novel zero-shot index dubbed Zen-Score to rank the architectures.
Neural Architecture Search
Vocal Bursts Intensity Prediction
1st Place Solution to ECCV-TAO-2020: Detect and Represent Any Object for Tracking
We extend the classical tracking-by-detection paradigm to this tracking-any-object task.
Ranked #7 on
Multi-Object Tracking
on TAO
(using extra training data)
On w-Optimization of the Split Covariance Intersection Filter
The split covariance intersection filter (split CIF) is a useful tool for general data fusion and has the potential to be applied in a variety of engineering tasks.
Object Detection Made Simpler by Eliminating Heuristic NMS
On the COCO dataset, our simple design achieves superior performance compared to both the FCOS baseline detector with NMS post-processing and the recent end-to-end NMS-free detectors.
Robust Representation Learning with Feedback for Single Image Deraining
Unlike existing image deraining methods that embed low-quality features into the model directly, we replace low-quality features by latent high-quality features.
A linearized framework and a new benchmark for model selection for fine-tuning
Since all model selection algorithms in the literature have been tested on different use-cases and never compared directly, we introduce a new comprehensive benchmark for model selection comprising of: i) A model zoo of single and multi-domain models, and ii) Many target tasks.
Zen-NAS: A Zero-Shot NAS for High-Performance Deep Image Recognition
Comparing with previous NAS methods, the proposed Zen-NAS is magnitude times faster on multiple server-side and mobile-side GPU platforms with state-of-the-art accuracy on ImageNet.
Ranked #2 on
Neural Architecture Search
on ImageNet
TransReID: Transformer-based Object Re-Identification
Extracting robust feature representation is one of the key challenges in object re-identification (ReID).
Ranked #1 on
Person Re-Identification
on Market-1501-C
MogFace: Towards a Deeper Appreciation on Face Detection
As a result, practical solutions on label assignment, scale-level data augmentation, and reducing false alarms are necessary for advancing face detectors.
Ranked #13 on
Face Detection
on WIDER Face (Easy)
Instant-Teaching: An End-to-End Semi-Supervised Object Detection Framework
To alleviate the confirmation bias problem and improve the quality of pseudo annotations, we further propose a co-rectify scheme based on Instant-Teaching, denoted as Instant-Teaching$^*$.
Ranked #12 on
Semi-Supervised Object Detection
on COCO 100% labeled data
(using extra training data)
AutoLoss-Zero: Searching Loss Functions from Scratch for Generic Tasks
However, the automatic design of loss functions for generic tasks with various evaluation metrics remains under-investigated.
PlenOctrees for Real-time Rendering of Neural Radiance Fields
We introduce a method to render Neural Radiance Fields (NeRFs) in real time using PlenOctrees, an octree-based 3D representation which supports view-dependent effects.
Equivariant Point Network for 3D Point Cloud Analysis
Features that are equivariant to a larger group of symmetries have been shown to be more discriminative and powerful in recent studies.
Guided Training: A Simple Method for Single-channel Speaker Separation
Another way is to use an anchor speech, a short speech of the target speaker, to model the speaker identity.
Augmented Transformer with Adaptive Graph for Temporal Action Proposal Generation
Temporal action proposal generation (TAPG) is a fundamental and challenging task in video understanding, especially in temporal action detection.
A Theoretical Analysis of Learning with Noisily Labeled Data
Noisy labels are very common in deep supervised learning.
Spatiotemporal Entropy Model is All You Need for Learned Video Compression
The framework of dominant learned video compression methods is usually composed of motion prediction modules as well as motion vector and residual image compression modules, suffering from its complex structure and error propagation problem.
A Simple Baseline for Semi-supervised Semantic Segmentation with Strong Data Augmentation
Previous works [3, 27] fail to employ strong augmentation in pseudo label learning efficiently, as the large distribution change caused by strong augmentation harms the batch normalisation statistics.
Learning to Cluster Faces via Transformer
In this paper, we repurpose the well-known Transformer and introduce a Face Transformer for supervised face clustering.
Spatially Self-Paced Convolutional Networks for Change Detection in Heterogeneous Images
Change detection in heterogeneous remote sensing images is a challenging problem because it is hard to make a direct comparison in the original observation spaces, and most methods rely on a set of manually labeled samples.
Why Does Multi-Epoch Training Help?
Stochastic gradient descent (SGD) has become the most attractive optimization method in training large-scale deep neural networks due to its simplicity, low computational cost in each updating step, and good performance.
Importance Weighted Adversarial Discriminative Transfer for Anomaly Detection
In this way, an obvious gap can be produced between the distributions of normal and abnormal data in the target domain, therefore enabling the anomaly detection in the domain.
City-Scale Multi-Camera Vehicle Tracking Guided by Crossroad Zones
Multi-Target Multi-Camera Tracking has a wide range of applications and is the basis for many advanced inferences and predictions.
Maximizing Mutual Information Across Feature and Topology Views for Learning Graph Representations
Specifically, we first utilize a multi-view representation learning module to better capture both local and global information content across feature and topology views on graphs.
An Empirical Study of Vehicle Re-Identification on the AI City Challenge
We mainly focus on four points, i. e. training data, unsupervised domain-adaptive (UDA) training, post-processing, model ensembling in this challenge.
An Efficient Training Approach for Very Large Scale Face Recognition
This method adopts Dynamic Class Pool (DCP) for storing and updating the identities features dynamically, which could be regarded as a substitute for the FC layer.
Ranked #1 on
Face Verification
on IJB-C
(training dataset metric)
Unsupervised Visual Representation Learning by Online Constrained K-Means
Clustering is to assign each instance a pseudo label that will be used to learn representations in discrimination.
Ranked #5 on
Unsupervised Image Classification
on CIFAR-10
KVT: k-NN Attention for Boosting Vision Transformers
A key component in vision transformers is the fully-connected self-attention which is more powerful than CNNs in modelling long range dependencies.
SKFAC:Training Neural Networks with Faster Kronecker-Factored Approximate Curvature
For the fully connected layers, by utilizing the low-rank property of Kronecker factors of Fisher information matrix, our method only requires inverting a small matrix to approximate the curvature with desirable accuracy.
Task-Generic Hierarchical Human Motion Prior using VAEs
We demonstrate the effectiveness of our hierarchical motion variational autoencoder in a variety of tasks including video-based human pose estimation, motion completion from partial observations, and motion synthesis from sparse key-frames.
Ranked #4 on
Motion Synthesis
on LaFAN1
SKFAC: Training Neural Networks With Faster Kronecker-Factored Approximate Curvature
For the fully connected layers, by utilizing the low-rank property of Kronecker factors of Fisher information matrix, our method only requires inverting a small matrix to approximate the curvature with desirable accuracy.
Normalized Avatar Synthesis Using StyleGAN and Perceptual Refinement
Cutting-edge 3D face reconstruction methods use non-linear morphable face models combined with GAN-based decoders to capture the likeness and details of a person but fail to produce neutral head models with unshaded albedo textures which is critical for creating relightable and animation-friendly avatars for integration in virtual environments.
Graph Convolution for Re-ranking in Person Re-identification
In this work, we propose a graph-based re-ranking method to improve learned features while still keeping Euclidean distance as the similarity metric.
LIFE: A Generalizable Autodidactic Pipeline for 3D OCT-A Vessel Segmentation
We then construct the local intensity fusion encoder (LIFE) to map a given OCT-A volume and its LIF counterpart to a shared latent space.
A Cloud-Edge-Terminal Collaborative System for Temperature Measurement in COVID-19 Prevention
Then, a mobile detection model based on a multi-task cascaded convolutional network (MTCNN) is proposed to realize face alignment and mask detection on the RGB images.
Fine-Grained AutoAugmentation for Multi-Label Classification
Data augmentation is a commonly used approach to improving the generalization of deep learning models.
Exploring the Quality of GAN Generated Images for Person Re-Identification
Recently, GAN based method has demonstrated strong effectiveness in generating augmentation data for person re-identification (ReID), on account of its ability to bridge the gap between domains and enrich the data variety in feature space.
Digging into Uncertainty in Self-supervised Multi-view Stereo
Specially, the limitations can be categorized into two types: ambiguious supervision in foreground and invalid supervision in background.
Dash: Semi-Supervised Learning with Dynamic Thresholding
In this work we develop a simple yet powerful framework, whose key idea is to select a subset of training examples from the unlabeled data when performing existing SSL methods so that only the unlabeled examples with pseudo labels related to the labeled data will be used to train models.
Scaled ReLU Matters for Training Vision Transformers
In this paper, we further investigate this problem and extend the above conclusion: only early convolutions do not help for stable training, but the scaled ReLU operation in the \textit{convolutional stem} (\textit{conv-stem}) matters.
CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation
Along with the pseudo labels, a weight-sharing triple-branch transformer framework is proposed to apply self-attention and cross-attention for source/target feature learning and source-target domain alignment, respectively.
Ranked #3 on
Domain Adaptation
on Office-31
DisUnknown: Distilling Unknown Factors for Disentanglement Learning
In this paper, we adopt a general setting where all factors that are hard to label or identify are encapsulated as a single unknown factor.
Interpolation variable rate image compression
Compression standards have been used to reduce the cost of image storage and transmission for decades.
Unsupervised Cross-Modality Domain Adaptation for Segmenting Vestibular Schwannoma and Cochlea with Data Augmentation and Model Ensemble
To overcome this problem, domain adaptation is an effective way to leverage information from source domain to obtain accurate segmentations without requiring manual labels in target domain.
Text-based Person Search in Full Images via Semantic-Driven Proposal Generation
Finding target persons in full scene images with a query of text description has important practical applications in intelligent video surveillance. However, different from the real-world scenarios where the bounding boxes are not available, existing text-based person retrieval methods mainly focus on the cross modal matching between the query text descriptions and the gallery of cropped pedestrian images.
Unsupervised Domain Adaptation By Optimal Transportation Of Clusters Between Domains
Unsupervised domain adaptation (UDA) aims to transfer the knowledge from a labeled source domain to an unlabeled target domain.
NAS-Bench-Zero: A Large Scale Dataset for Understanding Zero-Shot Neural Architecture Search
Based on these new discoveries, we propose i) a novel hybrid zero-shot proxy which outperforms existing ones by a large margin and is transferable among popular search spaces; ii) a new index for better measuring the true performance of ZS-NAS proxies in constrained NAS.
HSVA: Hierarchical Semantic-Visual Adaptation for Zero-Shot Learning
Specifically, HSVA aligns the semantic and visual domains by adopting a hierarchical two-step adaptation, i. e., structure adaptation and distribution adaptation.
Topologically Consistent Multi-View Face Inference Using Volumetric Sampling
We propose ToFu, Topologically consistent Face from multi-view, a geometry inference framework that can produce topologically consistent meshes across facial identities and expressions using a volumetric representation instead of an explicit underlying 3DMM.
Self-Supervised Pre-Training for Transformer-Based Person Re-Identification
We first investigate self-supervised learning (SSL) methods with Vision Transformer (ViT) pretrained on unlabelled person images (the LUPerson dataset), and empirically find it significantly surpasses ImageNet supervised pre-training models on ReID tasks.
Ranked #1 on
Unsupervised Person Re-Identification
on Market-1501
(using extra training data)
Get Better 1 Pixel PCK: Ladder Scales Correspondence Flow Networks for Remote Sensing Image Matching in Higher Resolution
Percentage of image size is often used as the threshold of PCK.
Improved Fine-Tuning by Better Leveraging Pre-Training Data
The generalization result of using pre-training data shows that the excess risk bound on a target task can be improved when the appropriate pre-training data is included in fine-tuning.
MAE-DET: Revisiting Maximum Entropy Principle in Zero-Shot NAS for Efficient Object Detection
Recent researches attempt to reduce this cost by optimizing the backbone architecture with the help of Neural Architecture Search (NAS).
Ranked #88 on
Object Detection
on COCO minival
3D High-Quality Magnetic Resonance Image Restoration in Clinics Using Deep Learning
We also analyzed several down-sampling strategies based on the acceleration factor, including multiple combinations of in-plane and through-plane down-sampling, and developed a controllable and quantifiable motion artifact generation method.
TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentation
Unsupervised semantic segmentation aims to obtain high-level semantic representation on low-level visual features without manual annotations.
Ranked #2 on
Unsupervised Semantic Segmentation
on COCO-Stuff-171
(using extra training data)
Uni-Perceiver: Pre-training Unified Architecture for Generic Perception for Zero-shot and Few-shot Tasks
The model is pre-trained on several uni-modal and multi-modal tasks, and evaluated on a variety of downstream tasks, including novel tasks that did not appear in the pre-training stage.
TransZero: Attribute-guided Transformer for Zero-Shot Learning
Although some attention-based models have attempted to learn such region features in a single image, the transferability and discriminative attribute localization of visual features are typically neglected.
Design and Implementation of Real-Time Localization System (RTLS) based on UWB and TDoA Algorithm
The challenges of indoor localization include inadequate localization accuracy, unreasonable anchor deployment in complex scenarios, lack of stability, and high cost.
On the Dilution of Precision for Time Difference of Arrival with Station Deployment
The paper aims to reveal the relationship between the performance of moving object tracking algorithms and the tracking anchors (station) deployment.
Decoupling and Recoupling Spatiotemporal Representation for RGB-D-based Motion Recognition
Decoupling spatiotemporal representation refers to decomposing the spatial and temporal features into dimension-independent factors.
Ranked #1 on
Hand Gesture Recognition
on NVGesture
Watch Those Words: Video Falsification Detection Using Word-Conditioned Facial Motion
In today's era of digital misinformation, we are increasingly faced with new threats posed by video falsification techniques.
ELSA: Enhanced Local Self-Attention for Vision Transformer
Self-attention is powerful in modeling long-range dependencies, but it is weak in local finer-level feature learning.
Ranked #46 on
Semantic Segmentation
on ADE20K val
Graph Neural Networks for Double-Strand DNA Breaks Prediction
Double-strand DNA breaks (DSBs) are a form of DNA damage that can cause abnormal chromosomal rearrangements.
CrossMoDA 2021 challenge: Benchmark of Cross-Modality Domain Adaptation techniques for Vestibular Schwannoma and Cochlea Segmentation
The aim was to automatically perform unilateral VS and bilateral cochlea segmentation on hrT2 as provided in the testing set (N=137).
Studying Popular Open Source Machine Learning Libraries and Their Cross-Ecosystem Bindings
Our study shows that the vast majority of the studied bindings cover only a small portion of the source library releases, and the delay for receiving support for a source library release is large.
Image-to-Video Re-Identification via Mutual Discriminative Knowledge Transfer
During the KD process, the TCL loss transfers the local structure, exploits the higher order information, and mitigates the misalignment of the heterogeneous output of teacher and student networks.
GiraffeDet: A Heavy-Neck Paradigm for Object Detection
This heavy-backbone design paradigm is mostly due to the historical legacy when transferring image recognition models to object detection rather than an end-to-end optimized design for object detection.
On Representation Learning with Feedback
This note complements the author's recent paper "Robust representation learning with feedback for single image deraining" by providing heuristically theoretical explanations on the mechanism of representation learning with feedback, namely an essential merit of the works presented in this recent article.
ModDrop++: A Dynamic Filter Network with Intra-subject Co-training for Multiple Sclerosis Lesion Segmentation with Missing Modalities
Previously, a training strategy termed Modality Dropout (ModDrop) has been applied to MS lesion segmentation to achieve the state-of-the-art performance with missing modality.
PMAL: Open Set Recognition via Robust Prototype Mining
Accordingly, the embedding space can be better optimized to discriminate therein the predefined classes and between known and unknowns.
EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation
The 2D-3D coordinates and corresponding weights are treated as intermediate variables learned by minimizing the KL divergence between the predicted and target pose distribution.
Ranked #6 on
6D Pose Estimation using RGB
on LineMOD
Task Adaptive Parameter Sharing for Multi-Task Learning
TAPS solves a joint optimization problem which determines which layers to share with the base model and the value of the task-specific weights.
Learning to Listen: Modeling Non-Deterministic Dyadic Facial Motion
We present a framework for modeling interactional communication in dyadic conversations: given multimodal inputs of a speaker, we autoregressively output multiple possibilities of corresponding listener motion.
Multi-view Point Cloud Registration based on Evolutionary Multitasking with Bi-Channel Knowledge Sharing Mechanism
The modeling of multi-view point cloud registration as multi-task optimization are twofold.
Joint learning of object graph and relation graph for visual question answering
Modeling visual question answering(VQA) through scene graphs can significantly improve the reasoning accuracy and interpretability.
NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results
The aim was to design a network for single image super-resolution that achieved improvement of efficiency measured according to several metrics including runtime, parameters, FLOPs, activations, and memory consumption while at least maintaining the PSNR of 29. 00dB on DIV2K validation set.
Unsupervised Representation Learning for 3D MRI Super Resolution with Degradation Adaptation
However, the difference in degradation representations between synthetic and authentic LR images suppresses the quality of SR images reconstructed from authentic LR images.
An Empirical Study on Distribution Shift Robustness From the Perspective of Pre-Training and Data Augmentation
With our empirical result obtained from 1, 330 models, we provide the following main observations: 1) ERM combined with data augmentation can achieve state-of-the-art performance if we choose a proper pre-trained model respecting the data property; 2) specialized algorithms further improve the robustness on top of ERM when handling a specific type of distribution shift, e. g., GroupDRO for spurious correlation and CORAL for large-scale out-of-distribution data; 3) Comparing different pre-training modes, architectures and data sizes, we provide novel observations about pre-training on distribution shift, which sheds light on designing or selecting pre-training strategy for different kinds of distribution shifts.
SwinVRNN: A Data-Driven Ensemble Forecasting Model via Learned Distribution Perturbation
We also compare four categories of perturbation methods for ensemble forecasting, i. e. fixed distribution perturbation, learned distribution perturbation, MC dropout, and multi model ensemble.
Point RCNN: An Angle-Free Framework for Rotated Object Detection
To tackle this problem, we propose a purely angle-free framework for rotated object detection, called Point RCNN, which mainly consists of PointRPN and PointReg.
