Ping Luo

Latest

• GenTron: Diffusion Transformers for Image and Video Generation
• RegionGPT: Towards Region Understanding Vision Language Model
• SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution
• MotionCtrl: A Unified and Flexible Motion Controller for Video Generation
• Part123: Part-aware 3D Reconstruction from a Single-view Image
• OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models
• PixArt-alpha: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis
• Vdt: General-purpose video diffusion transformers via mask modeling
• Embodiedgpt: Vision-language pre-training via embodied chain of thought
• Flow-Based Feature Fusion for Vehicle-Infrastructure Cooperative 3D Object Detection
• Foundation Model is Efficient Multimodal Multitask Model Selector
• Raphael: Text-to-image generation via large mixture of diffusion paths
• Visionllm: Large language model is also an open-ended decoder for vision-centric tasks
• Diffusiondet: Diffusion model for object detection
• Chipformer: Transferable chip placement via offline decision transformer
• pi-Tuning: Transferring Multimodal Foundation Models with Optimal Multi-task Interpolation
• Compression of Generative Pre-trained Language Models via Quantization
• CtrlFormer: Learning Transferable State Representation for Visual Control via Transformer
• Flow-based Recurrent Belief State Learning for POMDPs
• VLMixer: Unpaired Vision-Language Pre-training via Cross-Modal CutMix
• Bridging Video-text Retrieval with Multiple Choice Questions
• DanceTrack: Multi-Object Tracking in Uniform Appearance and Diverse Motion
• Language as Queries for Referring Video Object Segmentation
• Not All Tokens Are Equal: Human-centric Visual Analysis via Token Clustering Transformer
• Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers
• RestoreFormer: High-Quality Blind Face Restoration from Undegraded Key-Value Pairs
• Scale-Equivalent Distillation for Semi-Supervised Object Detection
• Don’t Touch What Matters: Task-Aware Lipschitz Data Augmentation for Visual Reinforcement Learning
• CycleMLP: A MLP-like Architecture for Dense Prediction
• Dynamic Token Normalization improves Vision Transformers
• Learning Versatile Neural Architectures by Propagating Network Codes
• Objects in Semantic Topology
• Pseudo-Labeled Auto-Curriculum Learning for Semi-Supervised Keypoint Localization
• Towards Ultra-Resolution Neural Style Transfer via Thumbnail Instance Normalization
• Adversarial Robustness for Unsupervised Domain Adaptation
• An Empirical Investigation of Representation Learning for Imitation
• Bringing Events Into Video Deblurring With Non-Consecutively Blurry Frames
• Compressed Video Contrastive Learning
• DetCo: Unsupervised Contrastive Learning for Object Detection
• Dynamic Visual Reasoning by Learning Differentiable Physics Models from Video and Language
• End-to-End Dense Video Captioning with Parallel Decoding
• Model-Based Reinforcement Learning via Imagination with Derived Memory
• Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
• Rethinking the Pruning Criteria for Convolutional Neural Network
• Revitalizing CNN Attention via Transformers in Self-Supervised Visual Representation Learning
• SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
• STAR: A Semantic-aware Transformer for Real-time Image Enhancement
• Watch Only Once: An End-to-End Video Action Detection Framework
• RelativeNAS: Relative Neural Architecture Search via Slow-Fast Learning
• Multi-frame Collaboration for Effective Endoscopic Video Polyp Detection via Spatial-Temporal Feature Transformation
• A Unified Multi-Scenario Attacking Network for Visual Object Tracking
• Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolution
• Disentangled Cycle Consistency for Highly-Realistic Virtual Try-On
• Do 2D GANs Know 3D Shape? Unsupervised 3D Shape Reconstruction from 2D Image GANs
• HR-NAS: Searching Efficient High-Resolution Neural Architectures With Lightweight Transformers
• Multi-Compound Transformer for Accurate Biomedical Image Segmentation
• Parser-Free Virtual Try-On via Distilling Appearance Flows
• PolarMask++: Enhanced Polar Representation for Single-Shot Instance Segmentation and Beyond
• Segmenting Transparent Objects in the Wild with Transformer
• Sparse R-CNN: End-to-End Object Detection With Learnable Proposals
• ViPNAS: Efficient Video Pose Estimation via Neural Architecture Search
• VPIPE: A Virtualized Acceleration System for Achieving Efficient and Scalable Pipeline Parallel DNN Training
• What Makes for End-to-End Object Detection?
• When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarks
• Switchable Normalization for Learning-to-Normalize Deep Representation
• UXNet: Searching Multi-level Feature Aggregation for 3D Medical Image Segmentation
• Recruitment and Opportunities
• 3D Human Mesh Regression with Dense Correspondence
• AdaX: Adaptive Gradient Descent with Exponential Long Term Memory
• Channel equilibrium networks for learning deep representation
• Convolution-Weight-Distribution Assumption: Rethinking the Criteria of Channel Pruning
• Differentiable Hierarchical Graph Grouping for Multi-Person Pose Estimation
• Domain-Adaptive Few-Shot Learning
• Every Frame Counts: Joint Learning of Video Segmentation and Optical Flow.
• Exemplar Normalization for Learning Deep Representation
• Exploiting Deep Generative Prior for Versatile Image Restoration and Manipulation
• How Does BN Increase Collapsed Neural Network Filters?
• Learning a Reinforced Agent for Flexible Exposure Bracketing Selection
• Learning depth-guided convolutions for monocular 3d object detection
• Maskgan: Towards diverse and interactive facial image manipulation
• Online Knowledge Distillation via Collaborative Learning
• Polarmask: Single shot instance segmentation with polar representation
• Segmenting Transparent Objects in the Wild
• Towards Photo-Realistic Virtual Try-On by Adaptively Generating-Preserving Image Content
• Whole-Body Human Pose Estimation in the Wild
• CamNet: Coarse-to-Fine Retrieval for Camera Re-Localization
• Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid
• Differentiable Learning-to-Group Channels via Groupable Convolutional Neural Networks
• Deep Self-Learning From Noisy Labels
• Once a MAN: Towards Multi-Target Attack via Learning Multi-Target Adversarial Network Once
• Vision-Infused Deep Audio Inpainting
• Switchable Whitening for Deep Representation Learning
• Differentiable Dynamic Normalization for Learning Deep Representation
• Understanding Normalization in Deep Learning
• DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images
• Differentiable learning-to-normalize via switchable normalization
• Do Normalization Layers in a Deep ConvNet Really Need to Be Distinct?
• Human Centric Visual Analysis with Deep Learning
• Towards Understanding Regularization in Batch Normalization
• Learning-to-Learn-to-Normalize: Algorithms, Applications and Theory
• 浅谈深度学习：归一化中的正则与泛化
• WIDER Face and Pedestrian Challenge 2018
• Batch Kalman Normalization: Towards Training Deep Neural Networks with Micro-Batches
• CUImage: A Neverending Learning Platform on a Convolutional Knowledge Graph of Billion Web Images
• Deep learning markov random field for semantic segmentation
• FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesis
• FaceID-GAN: Learning a Symmetry Three-Player GAN for Identity-Preserving Face Synthesis
• Faceness-net: Face detection through deep facial part responses
• From facial expression recognition to interpersonal relation prediction
• Kalman Normalization: Normalizing Internal Representations Across Network Layers
• Mix-and-match tuning for self-supervised semantic segmentation
• SCAN: Self-and-Collaborative Attention Network for Video Person Re-identification
• Scheduling Large-scale Distributed Training via Reinforcement Learning
• Spatial as deep: Spatial cnn for traffic scene understanding
• Talking Face Generation by Adversarially Disentangled Audio-Visual Representation
• Temporal Sequence Distillation: Towards Few-Frame Action Recognition in Videos
• Two at once: Enhancing learning and generalization capacities via ibn-net
• Deep Dual Learning for Semantic Image Segmentation
• Deep Learning Face Attributes for Detection and Alignment
• DeepID-Net: Object Detection with Deformable Part Based Convolutional Neural Networks
• EigenNet: Towards Fast and Structural Learning of Deep Neural Networks
• Learning deep architectures via generalized whitened neural networks
• Learning object interactions and descriptions for semantic image segmentation
• Not all pixels are equal: Difficulty-aware semantic segmentation via deep layer cascade
• Unconstrained fashion landmark detection via hierarchical recurrent transformer networks
• Video Classification via Relational Feature Encoding Networks
• Video object segmentation with re-identification
• Clothes Co-Parsing via Joint Image Segmentation and Labeling with Application to Clothing Retrieval
• Deepfashion: Powering robust clothes recognition and retrieval with rich annotations
• Face Model Compression by Distilling Knowledge from Neurons
• Fashion Landmark Detection in the Wild
• Joint face representation adaptation and clustering in videos
• Learning Compositional Shape Models of Multiple Distance Metrics by Information Projection.
• Learning deep representation for face alignment with auxiliary attributes
• Wider face: A face detection benchmark
• A large-scale car dataset for fine-grained categorization and verification
• Deep Learning Face Attributes in the Wild
• Deep Learning Strong Parts for Pedestrian Detection
• Deep Representation Learning with Target Coding.
• Deepid-net: Deformable deep convolutional neural networks for object detection
• From facial parts responses to face detection: A deep learning approach
• Learning social relation traits from face images
• Learning to recognize pedestrian attribute
• Pedestrian detection aided by deep learning semantic tasks
• Semantic image segmentation via deep parsing network
• Clothing Co-Parsing by Joint Image Segmentation and Labeling
• Deep learning multi-view representation for face recognition
• Deepid-net: multi-stage and deformable deep convolutional neural networks for object detection
• Facial landmark detection by deep multi-task learning
• Learning and Transferring Multi-task Deep Representation for Face Alignment
• Multi-view perceptron: a deep model for learning face identity and view representations
• Pedestrian attribute recognition at far distance
• Recover canonical-view faces in the wild with deep neural networks
• Switchable deep network for pedestrian detection
• A deep sum-product architecture for robust facial attributes analysis
• Deep learning identity-preserving face space
• Pedestrian parsing via deep decompositional network
• Hierarchical face parsing via deep learning
• Joint semantic segmentation by searching for compatible-competitive references
• Representing and recognizing objects with massive local image patches
• A Discriminative Model for Object Representation and Detection via Sparse Features
• Learning shape detector by quantizing curve segments with multiple distance metrics
• Semantics-driven portrait cartoon stylization
• Hierarchical 3D perception from a single image