Search Results for author:
Found 64 papers, 23 papers with code
Structural Supervision for Word Alignment and Machine Translation
Syntactic structure has long been argued to be potentially useful for enforcing accurate word alignment and improving generalization performance of machine translation.
From Knowledge Distillation to Self-Knowledge Distillation: A Unified Approach with Normalized Loss and Customized Soft Labels
We decompose the KD loss and find the non-target loss from it forces the student's non-target logits to match the teacher's, but the sum of the two non-target logits is different, preventing them from being identical.
Make Encoder Great Again in 3D GAN Inversion through Geometry and Occlusion-Aware Encoding
We leverage the inherent properties of EG3D's latent space to design a discriminator and a background depth regularization.
HiFace: High-Fidelity 3D Face Reconstruction by Learning Static and Dynamic Details
3D Morphable Models (3DMMs) demonstrate great potential for reconstructing faithful and animatable 3D facial surfaces from a single image.
Architecture, Dataset and Model-Scale Agnostic Data-free Meta-Learning
The goal of data-free meta-learning is to learn useful prior knowledge from a collection of pre-trained models without accessing their training data.
HMC: Hierarchical Mesh Coarsening for Skeleton-free Motion Retargeting
We present a simple yet effective method for skeleton-free motion retargeting.
TextIR: A Simple Framework for Text-based Editable Image Restoration
We use the text-image feature compatibility of the CLIP to alleviate the difficulty of fusing text and image features.
Rethink Long-tailed Recognition with Vision Transforms
Hence, we propose to adopt unsupervised learning to utilize long-tailed data.
UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers
Real-world data contains a vast amount of multimodal information, among which vision and language are the two most representative modalities.
ITstyler: Image-optimized Text-based Style Transfer
Text-based style transfer is a newly-emerging research topic that uses text information instead of style image to guide the transfer process, significantly extending the application scenario of style transfer.
Towards Arbitrary Text-driven Image Manipulation via Space Alignment
Then, the text input can be directly accessed into the StyleGAN space and be used to find the semantic shift according to the text description.
Truncate-Split-Contrast: A Framework for Learning from Mislabeled Videos
Thanks to Noise Contrastive Learning, the average classification accuracy improvement on Mini-Kinetics and Sth-Sth-V1 is over 1. 6\%.
Biomedical image analysis competitions: The state of current participation practice
Of these, 84% were based on standard architectures.
Learning Imbalanced Data with Vision Transformers
In this paper, we systematically investigate the ViTs' performance in LTR and propose LiVT to train ViTs from scratch only with LT data.
Ranked #5 on
Long-tail Learning
on CIFAR-10-LT (ρ=10)
High-fidelity Facial Avatar Reconstruction from Monocular Video with Generative Priors
Compared with existing works, we obtain superior novel view synthesis results and faithfully face reenactment performance.
Learning with Noisily-labeled Class-imbalanced Data
Specifically, RCAL works with the representations extracted by unsupervised contrastive learning.
Uni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks
In this paper, we propose Uni-Perceiver v2, which is the first generalist model capable of handling major large-scale vision and vision-language tasks with competitive performance.
One Model to Edit Them All: Free-Form Text-Driven Image Manipulation with Semantic Modulations
Based on the visual latent space of StyleGAN[21] and text embedding space of CLIP[34], studies focus on how to map these two latent spaces for text-driven attribute manipulations.
Darwinian Model Upgrades: Model Evolving with Selective Compatibility
The traditional model upgrading paradigm for retrieval requires recomputing all gallery embeddings before deploying the new model (dubbed as "backfilling"), which is quite expensive and time-consuming considering billions of instances in industrial applications.
Tackling Instance-Dependent Label Noise with Dynamic Distribution Calibration
In this paper, to address the distribution shift in learning with instance-dependent label noise, a dynamic distribution-calibration strategy is adopted.
Super-Resolution by Predicting Offsets: An Ultra-Efficient Super-Resolution Network for Rasterized Images
We finally use a guided fusion operation to integrate the sharp edges generated by the network and flat areas by the interpolation method to get the final SR image.
DeViT: Deformed Vision Transformers in Video Inpainting
This paper proposes a novel video inpainting method.
DCE: Offline Reinforcement Learning With Double Conservative Estimates
Previous conservative estimation methods are usually difficult to avoid the impact of OOD actions on Q-value estimates.
ViTKD: Practical Guidelines for ViT feature knowledge distillation
In this paper, we explore the way of feature-based distillation for ViT.
SWEM: Towards Real-Time Video Object Segmentation with Sequential Weighted Expectation-Maximization
Matching-based methods, especially those based on space-time memory, are significantly ahead of other solutions in semi-supervised video object segmentation (VOS).
Ranked #4 on
Semi-Supervised Video Object Segmentation
on MOSE
Semantic Segmentation
Semi-Supervised Video Object Segmentation
+1
Rethinking Knowledge Distillation via Cross-Entropy
Furthermore, we smooth students' target output to treat it as the soft target for training without teachers and propose a teacher-free new KD loss (tf-NKD).
HyP$^2$ Loss: Beyond Hypersphere Metric Space for Multi-label Image Retrieval
Image retrieval has become an increasingly appealing technique with broad multimedia application prospects, where deep hashing serves as the dominant branch towards low storage and efficient retrieval.
Improving the Latent Space of Image Style Transfer
The content contrastive loss enables the encoder to retain more available details.
Masked Generative Distillation
The current distillation algorithm usually improves students' performance by imitating the output of the teacher.
Privacy-Preserving Model Upgrades with Bidirectional Compatible Training in Image Retrieval
The task of privacy-preserving model upgrades in image retrieval desires to reap the benefits of rapidly evolving new models without accessing the raw gallery images.
REALY: Rethinking the Evaluation of 3D Face Reconstruction
The evaluation of 3D face reconstruction results typically relies on a rigid shape alignment between the estimated 3D model and the ground-truth scan.
Towards Universal Backward-Compatible Representation Learning
The task of backward-compatible representation learning is therefore introduced to support backfill-free model upgrades, where the new query features are interoperable with the old gallery features.
Hot-Refresh Model Upgrades with Regression-Alleviating Compatible Training in Image Retrieval
In contrast, hot-refresh model upgrades deploy the new model immediately and then gradually improve the retrieval accuracy by backfilling the gallery on-the-fly.
HHF: Hashing-guided Hinge Function for Deep Hashing Retrieval
Deep hashing has shown promising performance in large-scale image retrieval.
Semantic-Sparse Colorization Network for Deep Exemplar-based Colorization
To address these two problems, we propose Semantic-Sparse Colorization Network (SSCN) to transfer both the global image style and detailed semantic-related colors to the gray-scale image in a coarse-to-fine manner.
StrokeNet: Stroke Assisted and Hierarchical Graph Reasoning Networks
Scene text detection is still a challenging task, as there may be extremely small or low-resolution strokes, and close or arbitrary-shaped texts.
Focal and Global Knowledge Distillation for Detectors
Global distillation rebuilds the relation between different pixels and transfers it from teachers to students, compensating for missing global information in focal distillation.
Ranked #1 on
Knowledge Distillation
on COCO
Towards Calibrated Model for Long-Tailed Visual Recognition from Prior Perspective
Real-world data universally confronts a severe class-imbalance problem and exhibits a long-tailed distribution, i. e., most labels are associated with limited instances.
Ranked #18 on
Long-tail Learning
on CIFAR-100-LT (ρ=10)
MoDeRNN: Towards Fine-grained Motion Details for Spatiotemporal Predictive Learning
We carefully design Detail Context Block (DCB) to extract fine-grained details and improve the isolated correlation between upper context state and current input state.
BCDR: Betweenness Centrality-based Distance Resampling for Graph Shortest Distance Embedding
Second, we perform Distance Resampling (DR) from original walk paths before maximum likelihood optimization instead of the PMI-based optimization and prove that this strategy preserves distance relation with respect to any calibrated node via steering optimization objective to reconstruct a global distance matrix.
Superior Performance with Diversified Strategic Control in FPS Games Using General Reinforcement Learning
We provide comprehensive analysis and experiments to elaborate the effect of each component in affecting the agent performance, and demonstrate that the proposed and adopted techniques are important to achieve superior performance in general end-to-end FPS games.
Bootstrapped Hindsight Experience replay with Counterintuitive Prioritization
In this paper, inspired by Bootstrapped DQN, we use multiple heads in DDPG and take advantage of the diversity and uncertainty among multiple heads to improve the data efficiency with relabeled goals.
Hot-Refresh Model Upgrades with Regression-Free Compatible Training in Image Retrieval
In contrast, hot-refresh model upgrades deploy the new model immediately and then gradually improve the retrieval accuracy by backfilling the gallery on-the-fly.
Deep Open Snake Tracker for Vessel Tracing
Vessel tracing by modeling vascular structures in 3D medical images with centerlines and radii can provide useful information for vascular health.
CMS-LSTM: Context Embedding and Multi-Scale Spatiotemporal Expression LSTM for Predictive Learning
To tackle the increasing ambiguity during forecasting, we design CMS-LSTM to focus on context correlations and multi-scale spatiotemporal flow with details on fine-grained locals, containing two elaborate designed blocks: Context Embedding (CE) and Spatiotemporal Expression (SE) blocks.
A Self-Supervised Feature Map Augmentation (FMA) Loss and Combined Augmentations Finetuning to Efficiently Improve the Robustness of CNNs
Deep neural networks are often not robust to semantically-irrelevant changes in the input.
Bridge the Gap: High-level Semantic Planning for Image Captioning
To bridge the gaps we propose a high-level semantic planning (HSP) mechanism that incorporates both a semantic reconstruction and an explicit order planning.
Reducing the Annotation Effort for Video Object Segmentation Datasets
We use a deep convolutional network to automatically create pseudo-labels on a pixel level from much cheaper bounding box annotations and investigate how far such pseudo-labels can carry us for training state-of-the-art VOS approaches.
A Simple Yet Effective Method for Video Temporal Grounding with Cross-Modality Attention
The task of language-guided video temporal grounding is to localize the particular video clip corresponding to a query sentence in an untrimmed video.
HOSE-Net: Higher Order Structure Embedded Network for Scene Graph Generation
Second, since learning too many context-specific classification subspaces can suffer from data sparsity issues, we propose a hierarchical semantic aggregation(HSA) module to reduces the number of subspaces by introducing higher order structural information.
Automated Intracranial Artery Labeling using a Graph Neural Network and Hierarchical Refinement
Automatically labeling intracranial arteries (ICA) with their anatomical names is beneficial for feature extraction and detailed analysis of intracranial vascular structures.
Self-Attention ConvLSTM for Spatiotemporal Prediction
To extract spatial features with both global and local dependencies, we introduce the self-attention mechanism into ConvLSTM.
Ranked #22 on
Video Prediction
on Moving MNIST
Using an ensemble color space model to tackle adversarial examples
Minute pixel changes in an image drastically change the prediction that the deep learning model makes.
StegColNet: Steganalysis based on an ensemble colorspace approach
Image steganography refers to the process of hiding information inside images.
Multi-Frame Content Integration with a Spatio-Temporal Attention Mechanism for Person Video Motion Transfer
Specifically, we transfer the motion of one person in a target video to another person in a source video, while preserving the appearance of the source person.
Fast Registration for cross-source point clouds by using weak regional affinity and pixel-wise refinement
Accurate and fast registration of cross-source 3D point clouds from different sensors is an emerged research problem in computer vision.
ColorNet: Investigating the importance of color spaces for image classification
These color images are taken as input in the form of RGB images and classification is done without modifying them.
Ranked #4 on
Image Classification
on SVHN
Efficient Multi-level Correlating for Visual Tracking
Correlation filter (CF) based tracking algorithms have demonstrated favorable performance recently.
FPGA-based Acceleration System for Visual Tracking
In order to improve the tracking speed and reduce the overall power consumption of visual tracking, this paper proposes a real-time visual tracking algorithm based on DSST(Discriminative Scale Space Tracking) approach.
Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments
In the NIPS 2017 Learning to Run challenge, participants were tasked with building a controller for a musculoskeletal model to make it run as fast as possible through an obstacle course.
Y-net: 3D intracranial artery segmentation using a convolutional autoencoder
Automated segmentation of intracranial arteries on magnetic resonance angiography (MRA) allows for quantification of cerebrovascular features, which provides tools for understanding aging and pathophysiological adaptations of the cerebrovascular system.
A coarse-to-fine algorithm for registration in 3D street-view cross-source point clouds
In this paper, different from previous ICP-based methods, and from a statistic view, we propose a effective coarse-to-fine algorithm to detect and register a small scale SFM point cloud in a large scale Lidar point cloud.
Learning Boltzmann Machine with EM-like Method
We also propose a new measure to assess the performance of Boltzmann machine as generative models of data, and its computational complexity is O(Rmn).
A Systematic Approach for Cross-source Point Cloud Registration by Preserving Macro and Micro Structures
We propose a systematic approach for registering cross-source point clouds.
