Search Results for author:
Found 340 papers, 138 papers with code
Instance-Aware Embedding for Point Cloud Instance Segmentation
However, these methods are unaware of the instance context and fail to realize the boundary and geometric information of an instance, which are critical to separate adjacent objects.
A Geometric Perspective on Diffusion Models
Recent years have witnessed significant progress in developing efficient training and fast sampling approaches for diffusion models.
StyleAvatar3D: Leveraging Image-Text Diffusion Models for High-Fidelity 3D Avatar Generation
The recent advancements in image-text diffusion models have stimulated research interest in large-scale 3D generative models.
Learning Conditional Attributes for Compositional Zero-Shot Learning
Compositional Zero-Shot Learning (CZSL) aims to train models to recognize novel compositional concepts based on learned concepts such as attribute-object combinations.
Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning
We first design a PEFT-aware pruning criterion, which utilizes the values and gradients of Low-Rank Adaption (LoRA), rather than the gradients of pre-trained parameters for importance estimation.
Matcher: Segment Anything with One Shot Using All-Purpose Feature Matching
Naively connecting the models results in unsatisfying performance, e. g., the models tend to generate matching outliers and false-positive mask fragments.
SegGPT: Segmenting Everything In Context
We unify various segmentation tasks into a generalist in-context learning framework that accommodates different kinds of segmentation data by transforming them into the same format of images.
Ranked #1 on
Few-Shot Semantic Segmentation
on PASCAL-5i (5-Shot)
Zero-Shot Video Editing Using Off-The-Shelf Image Diffusion Models
Our vid2vid-zero leverages off-the-shelf image diffusion models, and doesn't require training on any video.
Zolly: Zoom Focal Length Correctly for Perspective-Distorted Human Mesh Reconstruction
As it is hard to calibrate single-view RGB images in the wild, existing 3D human mesh reconstruction (3DHMR) methods either use a constant large focal length or estimate one based on the background environment context, which can not tackle the problem of the torso, limb, hand or face distortion caused by perspective camera projection when the camera is close to the human body.
DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Models
In contrast, synthetic data can be freely available using a generative model (e. g., DALL-E, Stable Diffusion).
Background Matters: Enhancing Out-of-distribution Detection with Domain Features
To this end, we propose a novel generic framework that can learn the domain features from the ID training samples by a dense prediction approach, with which different existing semantic-feature-based OOD detection methods can be seamlessly combined to jointly learn the in-distribution features from both the semantic and domain dimensions.
Traffic Scene Parsing through the TSP6K Dataset
Traffic scene parsing is one of the most important tasks to achieve intelligent cities.
A Survey on Efficient Training of Transformers
Recent advances in Transformers have come with a huge requirement on computing resources, highlighting the importance of developing efficient training techniques to make Transformer training faster, at lower cost, and to higher accuracy by the efficient use of computation and memory resources.
SPTS v2: Single-Point Scene Text Spotting
End-to-end scene text spotting has made significant progress due to its intrinsic synergy between text detection and recognition.
Images Speak in Images: A Generalist Painter for In-Context Visual Learning
In this work, we present Painter, a generalist model which addresses these obstacles with an "image"-centric solution, that is, to redefine the output of core vision tasks as images, and specify task prompts as also images.
Ranked #6 on
Personalized Segmentation
on PerSeg
FoPro: Few-Shot Guided Robust Webly-Supervised Prototypical Learning
Most existing methods focus on learning noise-robust models from web images while neglecting the performance drop caused by the differences between web domain and real-world domain.
Learning from partially labeled data for multi-organ and tumor segmentation
To address this, we propose a Transformer based dynamic on-demand network (TransDoDNet) that learns to segment organs and tumors on multiple partially labeled datasets.
SC-DepthV3: Robust Self-supervised Monocular Depth Estimation for Dynamic Scenes
Self-supervised monocular depth estimation has shown impressive results in static scenes.
Indoor Monocular Depth Estimation
Monocular Depth Estimation
+1
Hierarchical Normalization for Robust Monocular Depth Estimation
In this paper, we address monocular depth estimation with deep neural networks.
Adv-Attribute: Inconspicuous and Transferable Adversarial Attack on Face Recognition
For face recognition attacks, existing methods typically generate the l_p-norm perturbations on pixels, however, resulting in low attack transferability and high vulnerability to denoising defense models.
SegViT: Semantic Segmentation with Plain Vision Transformers
We explore the capability of plain Vision Transformers (ViTs) for semantic segmentation and propose the SegVit.
Ranked #6 on
Semantic Segmentation
on PASCAL Context
Text-Adaptive Multiple Visual Prototype Matching for Video-Text Retrieval
To address this problem, we propose a Text-Adaptive Multiple Visual Prototype Matching model, which automatically captures multiple prototypes to describe a video by adaptive aggregation of video token features.
Multi-dataset Training of Transformers for Robust Action Recognition
We study the task of robust feature representations, aiming to generalize well on multiple datasets for action recognition.
Towards Accurate Reconstruction of 3D Scene Shape from A Single Monocular Image
To do so, we propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image, and then exploits 3D point cloud data to predict the depth shift and the camera's focal length that allow us to recover 3D scene shapes.
Towards Domain-agnostic Depth Completion
Our method leverages a data driven prior in the form of a single image depth prediction network trained on large-scale datasets, the output of which is used as an input to our model.
Real-time End-to-End Video Text Spotter with Contrastive Representation Learning
Our contributions are three-fold: 1) CoText simultaneously address the three tasks (e. g., text detection, tracking, recognition) in a real-time end-to-end trainable framework.
Efficient Decoder-free Object Detection with Transformers
A natural usage of ViTs in detection is to replace the CNN-based backbone with a transformer-based backbone, which is straightforward and effective, with the price of bringing considerable computation burden for inference.
Point-Teaching: Weakly Semi-Supervised Object Detection with Point Annotations
Point annotations are considerably more time-efficient than bounding box annotations.
Fully Convolutional One-Stage 3D Object Detection on LiDAR Range Images
In this work, we tackle this challenging issue with a novel range view projection mechanism, and for the first time demonstrate the benefits of fusing multi-frame point clouds for a range-view based detector.
Super Vision Transformer
Experimental results on ImageNet demonstrate that our SuperViT can considerably reduce the computational costs of ViT models with even performance increase.
PyramidCLIP: Hierarchical Feature Alignment for Vision-language Model Pretraining
Large-scale vision-language pre-training has achieved promising results on downstream tasks.
PointInst3D: Segmenting 3D Instances by Points
The current state-of-the-art methods in 3D instance segmentation typically involve a clustering step, despite the tendency towards heuristics, greedy algorithms, and a lack of robustness to the changes in data statistics.
TopFormer: Token Pyramid Transformer for Mobile Semantic Segmentation
Although vision transformers (ViTs) have achieved great success in computer vision, the heavy computational cost hampers their applications to dense prediction tasks such as semantic segmentation on mobile devices.
Improving Monocular Visual Odometry Using Learned Depth
The core of our framework is a monocular depth estimation module with a strong generalization capability for diverse scenes.
Catching Both Gray and Black Swans: Open-set Supervised Anomaly Detection
Despite most existing anomaly detection studies assume the availability of normal training samples only, a few labeled anomaly examples are often available in many real-world applications, such as defect samples identified during random quality inspection, lesion images confirmed by radiologists in daily medical screening, etc.
Ranked #2 on
supervised anomaly detection
on MVTec AD
(using extra training data)
End-to-End Video Text Spotting with Transformer
Recent video text spotting methods usually require the three-staged pipeline, i. e., detecting text in individual images, recognizing localized text, tracking text streams with post-processing to generate final results.
PointAttN: You Only Need Attention for Point Cloud Completion
To solve the problems, we leverage the cross-attention and self-attention mechanisms to design novel neural network for processing point cloud in a per-point manner to eliminate kNNs.
Training Protocol Matters: Towards Accurate Scene Text Recognition via Training Protocol Searching
The development of scene text recognition (STR) in the era of deep learning has been mainly focused on novel architectures of STR models.
Efficient Video Segmentation Models with Per-frame Inference
To this end, we perform inference at each frame.
FreeSOLO: Learning to Segment Objects without Annotations
FreeSOLO further demonstrates superiority as a strong pre-training method, outperforming state-of-the-art self-supervised pre-training methods by +9. 8% AP when fine-tuning instance segmentation with only 5% COCO masks.
Retrieval Augmented Classification for Long-Tail Visual Recognition
We introduce Retrieval Augmented Classification (RAC), a generic approach to augmenting standard image classification pipelines with an explicit retrieval module.
Ranked #3 on
Long-tail Learning
on iNaturalist 2018
The devil is in the labels: Semantic segmentation from sentences
The resulting merged semantic segmentation dataset of over 2 Million images enables training a model that achieves performance equal to that of state-of-the-art supervised methods on 7 benchmark datasets, despite not using any images therefrom.
Ranked #1 on
Semantic Segmentation
on WildDash
Towards 3D Scene Reconstruction from Locally Scale-Aligned Monocular Video Depth
However, in some video-based scenarios such as video depth estimation and 3D scene reconstruction from a video, the unknown scale and shift residing in per-frame prediction may cause the depth inconsistency.
Poseur: Direct Human Pose Regression with Transformers
We propose a direct, regression-based approach to 2D human pose estimation from single images.
Ranked #2 on
Keypoint Detection
on COCO
Boosting Robustness of Image Matting with Context Assembling and Strong Data Augmentation
Deep image matting methods have achieved increasingly better results on benchmarks (e. g., Composition-1k/alphamatting. com).
RigidFlow: Self-Supervised Scene Flow Learning on Point Clouds by Local Rigidity Prior
In this work, we focus on scene flow learning on point clouds in a self-supervised manner.
DENSE: Data-Free One-Shot Federated Learning
One-shot Federated Learning (FL) has recently emerged as a promising approach, which allows the central server to learn a model in a single communication round.
SPTS: Single-Point Text Spotting
For the first time, we demonstrate that training scene text spotting models can be achieved with an extremely low-cost annotation of a single-point for each instance.
NAS-FCOS: Efficient Search for Object Detection Architectures
Neural Architecture Search (NAS) has shown great potential in effectively reducing manual effort in network design by automatically discovering optimal architectures.
TSGB: Target-Selective Gradient Backprop for Probing CNN Visual Saliency
Inspired by those observations, we propose a novel visual saliency method, termed Target-Selective Gradient Backprop (TSGB), which leverages rectification operations to effectively emphasize target classes and further efficiently propagate the saliency to the image space, thereby generating target-selective and fine-grained saliency maps.
Meta Navigator: Search for a Good Adaptation Policy for Few-shot Learning
Inspired by the recent success in Automated Machine Learning literature (AutoML), in this paper, we present Meta Navigator, a framework that attempts to solve the aforementioned limitation in few-shot learning by seeking a higher-level strategy and proffer to automate the selection from various few-shot learning designs.
Explainable Deep Few-shot Anomaly Detection with Deviation Networks
Here, we study the problem of few-shot anomaly detection, in which we aim at using a few labeled anomaly examples to train sample-efficient discriminative detection models.
Ranked #3 on
supervised anomaly detection
on MVTec AD
(using extra training data)
Dynamic Neural Representational Decoders for High-Resolution Semantic Segmentation
This neural representation enables our decoder to leverage the smoothness prior in the semantic label space, and thus makes our decoder more efficient.
Dynamic Convolution for 3D Point Cloud Instance Segmentation
The proposed approach is proposal-free, and instead exploits a convolution process that adapts to the spatial and semantic characteristics of each instance.
SOLO: A Simple Framework for Instance Segmentation
Besides instance segmentation, our method yields state-of-the-art results in object detection (from our mask byproduct) and panoptic segmentation.
Learning Spatial-Semantic Relationship for Facial Attribute Recognition With Limited Labeled Data
First, three auxiliary tasks, consisting of a Patch Rotation Task (PRT), a Patch Segmentation Task (PST), and a Patch Classification Task (PCT), are jointly developed to learn the spatial-semantic relationship from large-scale unlabeled facial data.
FCPose: Fully Convolutional Multi-Person Pose Estimation with Dynamic Instance-Aware Convolutions
We propose a fully convolutional multi-person pose estimation framework using dynamic instance-aware convolutions, termed FCPose.
Unsupervised Scale-consistent Depth Learning from Video
We propose a monocular depth estimator SC-Depth, which requires only unlabelled videos for training and enables the scale-consistent prediction at inference time.
Ranked #38 on
Monocular Depth Estimation
on KITTI Eigen split
HCRF-Flow: Scene Flow from Point Clouds with Continuous High-order CRFs and Position-aware Flow Embedding
Scene flow in 3D point clouds plays an important role in understanding dynamic environments.
ABCNet v2: Adaptive Bezier-Curve Network for Real-time End-to-end Text Spotting
Previous methods can be roughly categorized into two groups: character-based and segmentation-based, which often require character-level annotations and/or complex post-processing due to the unstructured output.
PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text
By systematically comparing with existing scene text representations, we show that our kernel representation can not only describe arbitrarily-shaped text but also well distinguish adjacent text.
Twins: Revisiting the Design of Spatial Attention in Vision Transformers
Very recently, a variety of vision transformer architectures for dense prediction tasks have been proposed and they show that the design of spatial attention is critical to their success in these tasks.
Ranked #47 on
Semantic Segmentation
on ADE20K val
DisCo: Remedy Self-supervised Learning on Lightweight Models with Distilled Contrastive Learning
Specifically, we find the final embedding obtained by the mainstream SSL methods contains the most fruitful information, and propose to distill the final embedding to maximally transmit a teacher's knowledge to a lightweight model by constraining the last embedding of the student to be consistent with that of the teacher.
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.
Feature Decomposition and Reconstruction Learning for Effective Facial Expression Recognition
In this paper, we propose a novel Feature Decomposition and Reconstruction Learning (FDRL) method for effective facial expression recognition.
TFPose: Direct Human Pose Estimation with Transformers
We propose a human pose estimation framework that solves the task in the regression-based fashion.
Ranked #26 on
Pose Estimation
on MPII Human Pose
An Adversarial Human Pose Estimation Network Injected with Graph Structure
Because of the invisible human keypoints in images caused by illumination, occlusion and overlap, it is likely to produce unreasonable human pose prediction for most of the current human pose estimation methods.
Generic Perceptual Loss for Modeling Structured Output Dependencies
We hope that this simple, extended perceptual loss may serve as a generic structured-output loss that is applicable to most structured output learning tasks.
FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation
Experiments on both synthetic Sintel data and real-world KITTI datasets demonstrate the effectiveness of the proposed approach, which needs only 1/10 computation of comparable networks to achieve on par accuracy.
Virtual Normal: Enforcing Geometric Constraints for Accurate and Robust Depth Prediction
In this work, we show the importance of the high-order 3D geometric constraints for depth prediction.
CoTr: Efficiently Bridging CNN and Transformer for 3D Medical Image Segmentation
Convolutional neural networks (CNNs) have been the de facto standard for nowadays 3D medical image segmentation.
Conditional Positional Encodings for Vision Transformers
Built on PEG, we present Conditional Position encoding Vision Transformer (CPVT).
Instance and Panoptic Segmentation Using Conditional Convolutions
In the literature, top-performing instance segmentation methods typically follow the paradigm of Mask R-CNN and rely on ROI operations (typically ROIAlign) to attend to each instance.
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.
Multi-intersection Traffic Optimisation: A Benchmark Dataset and a Strong Baseline
To address these issues, in this work we carefully design our settings and propose a new dataset including both synthetic and real traffic data in more complex scenarios.
Single-path Bit Sharing for Automatic Loss-aware Model Compression
By jointly training the binary gates in conjunction with network parameters, the compression configurations of each layer can be automatically determined.
BV-Person: A Large-Scale Dataset for Bird-View Person Re-Identification
In this work we introduce a new ReID task, bird-view person ReID, which aims at searching for a person in a gallery of horizontal-view images with the query images taken from a bird's-eye view, i. e., an elevated view of an object from above.
Occluded Person Re-Identification With Single-Scale Global Representations
However, real-world ReID applications typically have highly diverse occlusions and involve a hybrid of occluded and non-occluded pedestrians.
Memory-Efficient Hierarchical Neural Architecture Search for Image Restoration
For the inner search space, we propose a layer-wise architecture sharing strategy (LWAS), resulting in more flexible architectures and better performance.
Diverse Knowledge Distillation for End-to-End Person Search
Person search aims to localize and identify a specific person from a gallery of images.
Learning to Recover 3D Scene Shape from a Single Image
Despite significant progress in monocular depth estimation in the wild, recent state-of-the-art methods cannot be used to recover accurate 3D scene shape due to an unknown depth shift induced by shift-invariant reconstruction losses used in mixed-data depth prediction training, and possible unknown camera focal length.
Ranked #1 on
Depth Estimation
on ScanNetV2
Hyperspectral Classification Based on Lightweight 3-D-CNN With Transfer Learning
In contrast to previous approaches, we do not impose restrictions over the source data sets, in which they do not have to be collected by the same sensors as the target data sets.
BoxInst: High-Performance Instance Segmentation with Box Annotations
We present a high-performance method that can achieve mask-level instance segmentation with only bounding-box annotations for training.
Box-supervised Instance Segmentation
Semantic Segmentation
+2
End-to-End Video Instance Segmentation with Transformers
Here, we propose a new video instance segmentation framework built upon Transformers, termed VisTR, which views the VIS task as a direct end-to-end parallel sequence decoding/prediction problem.
Ranked #23 on
Video Instance Segmentation
on YouTube-VIS validation
Fully Quantized Image Super-Resolution Networks
With the rising popularity of intelligent mobile devices, it is of great practical significance to develop accurate, realtime and energy-efficient image Super-Resolution (SR) inference methods.
Learning Affinity-Aware Upsampling for Deep Image Matting
By looking at existing upsampling operators from a unified mathematical perspective, we generalize them into a second-order form and introduce Affinity-Aware Upsampling (A2U) where upsampling kernels are generated using a light-weight lowrank bilinear model and are conditioned on second-order features.
Channel-wise Knowledge Distillation for Dense Prediction
Observing that in semantic segmentation, some layers' feature activations of each channel tend to encode saliency of scene categories (analogue to class activation mapping), we propose to align features channel-wise between the student and teacher networks.
DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution
Previous top-performing approaches for point cloud instance segmentation involve a bottom-up strategy, which often includes inefficient operations or complex pipelines, such as grouping over-segmented components, introducing additional steps for refining, or designing complicated loss functions.
PGL: Prior-Guided Local Self-supervised Learning for 3D Medical Image Segmentation
In this paper, we propose a PriorGuided Local (PGL) self-supervised model that learns the region-wise local consistency in the latent feature space.
Graph Attention Tracking
We propose to establish part-to-part correspondence between the target and the search region with a complete bipartite graph, and apply the graph attention mechanism to propagate target information from the template feature to the search feature.
Robust Data Hiding Using Inverse Gradient Attention
In deep data hiding models, to maximize the encoding capacity, each pixel of the cover image ought to be treated differently since they have different sensitivities w. r. t.
DoDNet: Learning to segment multi-organ and tumors from multiple partially labeled datasets
To address this, we propose a dynamic on-demand network (DoDNet) that learns to segment multiple organs and tumors on partially labeled datasets.
Unifying Instance and Panoptic Segmentation with Dynamic Rank-1 Convolutions
To our knowledge, DR1Mask is the first panoptic segmentation framework that exploits a shared feature map for both instance and semantic segmentation by considering both efficacy and efficiency.
Dense Contrastive Learning for Self-Supervised Visual Pre-Training
Compared to the baseline method MoCo-v2, our method introduces negligible computation overhead (only <1% slower), but demonstrates consistently superior performance when transferring to downstream dense prediction tasks including object detection, semantic segmentation and instance segmentation; and outperforms the state-of-the-art methods by a large margin.
Toward Deep Supervised Anomaly Detection: Reinforcement Learning from Partially Labeled Anomaly Data
We consider the problem of anomaly detection with a small set of partially labeled anomaly examples and a large-scale unlabeled dataset.
FATNN: Fast and Accurate Ternary Neural Networks
Ternary Neural Networks (TNNs) have received much attention due to being potentially orders of magnitude faster in inference, as well as more power efficient, than full-precision counterparts.
Representative Graph Neural Network
In this paper, we present a Representative Graph (RepGraph) layer to dynamically sample a few representative features, which dramatically reduces redundancy.
Pairwise Relation Learning for Semi-supervised Gland Segmentation
In this paper, we propose the pairwise relation-based semi-supervised (PRS^2) model for gland segmentation on histology images.
AE TextSpotter: Learning Visual and Linguistic Representation for Ambiguous Text Spotting
Unlike previous works that merely employed visual features for text detection, this work proposes a novel text spotter, named Ambiguity Eliminating Text Spotter (AE TextSpotter), which learns both visual and linguistic features to significantly reduce ambiguity in text detection.
Improving Generative Adversarial Networks with Local Coordinate Coding
In this paper, rather than sampling from the predefined prior distribution, we propose an LCCGAN model with local coordinate coding (LCC) to improve the performance of generating data.
Soft Expert Reward Learning for Vision-and-Language Navigation
In this paper, we introduce a Soft Expert Reward Learning (SERL) model to overcome the reward engineering designing and generalisation problems of the VLN task.
Weighing Counts: Sequential Crowd Counting by Reinforcement Learning
Inspired by scale weighing, we propose a novel 'counting scale' termed LibraNet where the count value is analogized by weight.
AQD: Towards Accurate Quantized Object Detection
Network quantization allows inference to be conducted using low-precision arithmetic for improved inference efficiency of deep neural networks on edge devices.
Deep Learning for Anomaly Detection: A Review
This paper surveys the research of deep anomaly detection with a comprehensive taxonomy, covering advancements in three high-level categories and 11 fine-grained categories of the methods.
FCOS: A simple and strong anchor-free object detector
In computer vision, object detection is one of most important tasks, which underpins a few instance-level recognition tasks and many downstream applications.
A Robust Attentional Framework for License Plate Recognition in the Wild
On the other hand, the 2D attentional based license plate recognizer with an Xception-based CNN encoder is capable of recognizing license plates with different patterns under various scenarios accurately and robustly.
Auto-Rectify Network for Unsupervised Indoor Depth Estimation
However, excellent results have mostly been obtained in street-scene driving scenarios, and such methods often fail in other settings, particularly indoor videos taken by handheld devices.
Ranked #42 on
Monocular Depth Estimation
on NYU-Depth V2
Scope Head for Accurate Localization in Object Detection
Existing anchor-based and anchor-free object detectors in multi-stage or one-stage pipelines have achieved very promising detection performance.
Scene Text Image Super-Resolution in the Wild
For example, it outperforms LapSRN by over 5% and 8%on the recognition accuracy of ASTER and CRNN.
BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
We propose to treat these spatial details and categorical semantics separately to achieve high accuracy and high efficiency for realtime semantic segmentation.
Ranked #1 on
Real-Time Semantic Segmentation
on COCO-Stuff
Context Prior for Scene Segmentation
Given an input image and corresponding ground truth, Affinity Loss constructs an ideal affinity map to supervise the learning of Context Prior.
Ranked #1 on
Scene Understanding
on ADE20K val
Segmenting Transparent Objects in the Wild
To address this important problem, this work proposes a large-scale dataset for transparent object segmentation, named Trans10K, consisting of 10, 428 images of real scenarios with carefully manual annotations, which are 10 times larger than the existing datasets.
Ranked #4 on
Semantic Segmentation
on Trans10K
Viral Pneumonia Screening on Chest X-ray Images Using Confidence-Aware Anomaly Detection
In this paper, we formulate the task of differentiating viral pneumonia from non-viral pneumonia and healthy controls into an one-class classification-based anomaly detection problem, and thus propose the confidence-aware anomaly detection (CAAD) model, which consists of a shared feature extractor, an anomaly detection module, and a confidence prediction module.
Mask Encoding for Single Shot Instance Segmentation
To date, instance segmentation is dominated by twostage methods, as pioneered by Mask R-CNN.
SOLOv2: Dynamic and Fast Instance Segmentation
Importantly, we take one step further by dynamically learning the mask head of the object segmenter such that the mask head is conditioned on the location.
Ranked #10 on
Real-time Instance Segmentation
on MSCOCO
DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learning
We employ the Earth Mover's Distance (EMD) as a metric to compute a structural distance between dense image representations to determine image relevance.
Self-trained Deep Ordinal Regression for End-to-End Video Anomaly Detection
Video anomaly detection is of critical practical importance to a variety of real applications because it allows human attention to be focused on events that are likely to be of interest, in spite of an otherwise overwhelming volume of video.
Conditional Convolutions for Instance Segmentation
We propose a simple yet effective instance segmentation framework, termed CondInst (conditional convolutions for instance segmentation).
Real-Time High-Performance Semantic Image Segmentation of Urban Street Scenes
Meanwhile, a Spatial detail-Preserving Network (SPN) with shallow convolutional layers is designed to generate high-resolution feature maps preserving the detailed spatial information.
Efficient Semantic Video Segmentation with Per-frame Inference
For semantic segmentation, most existing real-time deep models trained with each frame independently may produce inconsistent results for a video sequence.
Ranked #2 on
Video Semantic Segmentation
on CamVid
On the General Value of Evidence, and Bilingual Scene-Text Visual Question Answering
Visual Question Answering (VQA) methods have made incredible progress, but suffer from a failure to generalize.
ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network
Our contributions are three-fold: 1) For the first time, we adaptively fit arbitrarily-shaped text by a parameterized Bezier curve.
Joint Deep Learning of Facial Expression Synthesis and Recognition
Firstly, a facial expression synthesis generative adversarial network (FESGAN) is pre-trained to generate facial images with different facial expressions.
DiverseDepth: Affine-invariant Depth Prediction Using Diverse Data
Compared with previous learning objectives, i. e., learning metric depth or relative depth, we propose to learn the affine-invariant depth using our diverse dataset to ensure both generalization and high-quality geometric shapes of scenes.
Separating Content from Style Using Adversarial Learning for Recognizing Text in the Wild
Furthermore, to tackle the issue of lacking paired training samples, we design an interactive joint training scheme, which shares attention masks from the recognizer to the discriminator, and enables the discriminator to extract the features of each character for further adversarial training.
Memorizing Comprehensively to Learn Adaptively: Unsupervised Cross-Domain Person Re-ID with Multi-level Memory
Due to the lack of supervision in the target domain, it is crucial to identify the underlying similarity-and-dissimilarity relationships among the unlabelled samples in the target domain.
From Open Set to Closed Set: Supervised Spatial Divide-and-Conquer for Object Counting
Visual counting, a task that aims to estimate the number of objects from an image/video, is an open-set problem by nature, i. e., the number of population can vary in [0, inf) in theory.
Learning and Memorizing Representative Prototypes for 3D Point Cloud Semantic and Instance Segmentation
3D point cloud semantic and instance segmentation is crucial and fundamental for 3D scene understanding.
Ranked #21 on
3D Instance Segmentation
on ScanNet(v2)
(mAP @ 50 metric)
BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation
The proposed BlendMask can effectively predict dense per-pixel position-sensitive instance features with very few channels, and learn attention maps for each instance with merely one convolution layer, thus being fast in inference.
Ranked #13 on
Real-time Instance Segmentation
on MSCOCO
Ordered or Orderless: A Revisit for Video based Person Re-Identification
Specifically, with a diagnostic analysis, we show that the recurrent structure may not be effective to learn temporal dependencies than what we expected and implicitly yields an orderless representation.
Unsupervised Representation Learning by Predicting Random Distances
To enable unsupervised learning on those domains, in this work we propose to learn features without using any labelled data by training neural networks to predict data distances in a randomly projected space.
Exploring the Capacity of an Orderless Box Discretization Network for Multi-orientation Scene Text Detection
More importantly, based on OBD, we provide a detailed analysis of the impact of a collection of refinements, which may inspire others to build state-of-the-art text detectors.
Ranked #3 on
Scene Text Detection
on ICDAR 2017 MLT
To Balance or Not to Balance: A Simple-yet-Effective Approach for Learning with Long-Tailed Distributions
Such imbalanced distribution causes a great challenge for learning a deep neural network, which can be boiled down into a dilemma: on the one hand, we prefer to increase the exposure of tail class samples to avoid the excessive dominance of head classes in the classifier training.
SOLO: Segmenting Objects by Locations
We present a new, embarrassingly simple approach to instance segmentation in images.
Ranked #64 on
Instance Segmentation
on COCO test-dev
Unified Multifaceted Feature Learning for Person Re-Identification
The loss structures the augmented images resulted by the two types of image erasing in a two-level hierarchy and enforces multifaceted attention to different parts.
Deep Anomaly Detection with Deviation Networks
Instead of representation learning, our method fulfills an end-to-end learning of anomaly scores by a neural deviation learning, in which we leverage a few (e. g., multiple to dozens) labeled anomalies and a prior probability to enforce statistically significant deviations of the anomaly scores of anomalies from that of normal data objects in the upper tail.
Ranked #1 on
Network Intrusion Detection
on NB15-Backdoor
DirectPose: Direct End-to-End Multi-Person Pose Estimation
We propose the first direct end-to-end multi-person pose estimation framework, termed DirectPose.
Ranked #13 on
Keypoint Detection
on COCO test-dev
Multi-marginal Wasserstein GAN
Multiple marginal matching problem aims at learning mappings to match a source domain to multiple target domains and it has attracted great attention in many applications, such as multi-domain image translation.
Deep Weakly-supervised Anomaly Detection
To detect both seen and unseen anomalies, we introduce a novel deep weakly-supervised approach, namely Pairwise Relation prediction Network (PReNet), that learns pairwise relation features and anomaly scores by predicting the relation of any two randomly sampled training instances, in which the pairwise relation can be anomaly-anomaly, anomaly-unlabeled, or unlabeled-unlabeled.
Semi-supervised Anomaly Detection
supervised anomaly detection
+1
PolarMask: Single Shot Instance Segmentation with Polar Representation
In this paper, we introduce an anchor-box free and single shot instance segmentation method, which is conceptually simple, fully convolutional and can be used as a mask prediction module for instance segmentation, by easily embedding it into most off-the-shelf detection methods.
Ranked #97 on
Instance Segmentation
on COCO test-dev
Structured Binary Neural Networks for Image Recognition
Experiments on both classification, semantic segmentation and object detection tasks demonstrate the superior performance of the proposed methods over various quantized networks in the literature.
Memory-Efficient Hierarchical Neural Architecture Search for Image Denoising
We also present analysis on the architectures found by NAS.
Task-Aware Monocular Depth Estimation for 3D Object Detection
In this paper, we first analyse the data distributions and interaction of foreground and background, then propose the foreground-background separated monocular depth estimation (ForeSeE) method, to estimate the foreground depth and background depth using separate optimization objectives and depth decoders.
TextSR: Content-Aware Text Super-Resolution Guided by Recognition
Nonetheless, most of the previous methods may not work well in recognizing text with low resolution which is often seen in natural scene images.
Part-Guided Attention Learning for Vehicle Instance Retrieval
Finally, we aggregate the global appearance and part features to improve the feature performance further.
Auxiliary Learning for Deep Multi-task Learning
The most current methods can be categorized as either: (i) hard parameter sharing where a subset of the parameters is shared among tasks while other parameters are task-specific; or (ii) soft parameter sharing where all parameters are task-specific but they are jointly regularized.
Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video
To the best of our knowledge, this is the first work to show that deep networks trained using unlabelled monocular videos can predict globally scale-consistent camera trajectories over a long video sequence.
Ranked #43 on
Monocular Depth Estimation
on KITTI Eigen split
Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network
Recently, some methods have been proposed to tackle arbitrary-shaped text detection, but they rarely take the speed of the entire pipeline into consideration, which may fall short in practical applications. In this paper, we propose an efficient and accurate arbitrary-shaped text detector, termed Pixel Aggregation Network (PAN), which is equipped with a low computational-cost segmentation head and a learnable post-processing.
Ranked #6 on
Scene Text Detection
on SCUT-CTW1500
From Open Set to Closed Set: Counting Objects by Spatial Divide-and-Conquer
A dense region can always be divided until sub-region counts are within the previously observed closed set.
Ranked #3 on
Crowd Counting
on TRANCOS
Index Network
By viewing the indices as a function of the feature map, we introduce the concept of "learning to index", and present a novel index-guided encoder-decoder framework where indices are self-learned adaptively from data and are used to guide the downsampling and upsampling stages, without extra training supervision.
Ranked #2 on
Grayscale Image Denoising
on Set12 sigma30
MobileFAN: Transferring Deep Hidden Representation for Face Alignment
To this end, we propose an effective lightweight model, namely Mobile Face Alignment Network (MobileFAN), using a simple backbone MobileNetV2 as the encoder and three deconvolutional layers as the decoder.
Exploiting temporal consistency for real-time video depth estimation
The temporal consistency loss is combined with the spatial loss to update the model in an end-to-end fashion.
Ranked #5 on
Monocular Depth Estimation
on Mid-Air Dataset
Effective Training of Convolutional Neural Networks with Low-bitwidth Weights and Activations
Furthermore, we propose a second progressive quantization scheme which gradually decreases the bit-width from high-precision to low-precision during training.
Indices Matter: Learning to Index for Deep Image Matting
We show that existing upsampling operators can be unified with the notion of the index function.
Self-training with progressive augmentation for unsupervised cross-domain person re-identification
In this work, we develop a self-training method with progressive augmentation framework (PAST) to promote the model performance progressively on the target dataset.
Ranked #11 on
Unsupervised Domain Adaptation
on Market to Duke
V-PROM: A Benchmark for Visual Reasoning Using Visual Progressive Matrices
One of the primary challenges faced by deep learning is the degree to which current methods exploit superficial statistics and dataset bias, rather than learning to generalise over the specific representations they have experienced.