ICCV2023論文速遞（2023.8.15）！
最新成果demo展示：
ICCV 2023
Updated on : 15 Aug 2023
total number : 0
Group Pose: A Simple Baseline for End-to-End Multi-person Pose Estimation
論文/Paper: http://arxiv.org/pdf/2308.07313
代碼/Code: https://github.com/michel-liu/grouppose
Distance Matters For Improving Performance Estimation Under Covariate Shift
論文/Paper: http://arxiv.org/pdf/2308.07223
代碼/Code: https://github.com/melanibe/distance_matters_performance_estimation
HyperSparse Neural Networks: Shifting Exploration to Exploitation through Adaptive Regularization
論文/Paper: http://arxiv.org/pdf/2308.07163
代碼/Code: https://github.com/greenautoml4fas/hypersparse
DELO: Deep Evidential LiDAR Odometry using Partial Optimal Transport
論文/Paper: http://arxiv.org/pdf/2308.07153
代碼/Code: None
Diffusion Based Augmentation for Captioning and Retrieval in Cultural Heritage
論文/Paper: http://arxiv.org/pdf/2308.07151
代碼/Code: None
CTP: Towards Vision-Language Continual Pretraining via Compatible Momentum Contrast and Topology Preservation
論文/Paper: http://arxiv.org/pdf/2308.07146
代碼/Code: https://github.com/kevinlight831/ctp
SCSC: Spatial Cross-scale Convolution Module to Strengthen both CNNs and Transformers
論文/Paper: http://arxiv.org/pdf/2308.07110
代碼/Code: None
Masked Motion Predictors are Strong 3D Action Representation Learners
論文/Paper: http://arxiv.org/pdf/2308.07092
代碼/Code: https://github.com/maoyunyao/MAMP.
S3IM: Stochastic Structural SIMilarity and Its Unreasonable Effectiveness for Neural Fields
論文/Paper: http://arxiv.org/pdf/2308.07032
代碼/Code: https://github.com/Madaoer/S3IM.
ACTIVE: Towards Highly Transferable 3D Physical Camouflage for Universal and Robust Vehicle Evasion
論文/Paper: http://arxiv.org/pdf/2308.07009
代碼/Code: None
Global Features are All You Need for Image Retrieval and Reranking
論文/Paper: http://arxiv.org/pdf/2308.06954
代碼/Code: https://github.com/shihaoshao-gh/superglobal
Knowing Where to Focus: Event-aware Transformer for Video Grounding
論文/Paper: http://arxiv.org/pdf/2308.06947
代碼/Code: https://github.com/jinhyunj/eatr
Exploring Lightweight Hierarchical Vision Transformers for Efficient Visual Tracking
論文/Paper: http://arxiv.org/pdf/2308.06904
代碼/Code: https://github.com/kangben258/hit
Towards Open-Set Test-Time Adaptation Utilizing the Wisdom of Crowds in Entropy Minimization
論文/Paper: http://arxiv.org/pdf/2308.06879
代碼/Code: None
RMP-Loss: Regularizing Membrane Potential Distribution for Spiking Neural Networks
論文/Paper: http://arxiv.org/pdf/2308.06787
代碼/Code: None
Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning
論文/Paper: http://arxiv.org/pdf/2308.06777
代碼/Code: https://github.com/LiheYoung/ShrinkMatch
Dual Meta-Learning with Longitudinally Generalized Regularization for One-Shot Brain Tissue Segmentation Across the Human Lifespan
論文/Paper: http://arxiv.org/pdf/2308.06774
代碼/Code: https://github.com/ladderlab-xjtu/DuMeta.
AerialVLN: Vision-and-Language Navigation for UAVs
論文/Paper: http://arxiv.org/pdf/2308.06735
代碼/Code: https://github.com/AirVLN/AirVLN.
Compositional Feature Augmentation for Unbiased Scene Graph Generation
論文/Paper: http://arxiv.org/pdf/2308.06712
代碼/Code: None
Isomer: Isomerous Transformer for Zero-shot Video Object Segmentation
論文/Paper: http://arxiv.org/pdf/2308.06693
代碼/Code: https://github.com/dlut-yyc/isomer
Estimator Meets Equilibrium Perspective: A Rectified Straight Through Estimator for Binary Neural Networks Training
論文/Paper: http://arxiv.org/pdf/2308.06689
代碼/Code: https://github.com/dravenalg/reste
3DMOTFormer: Graph Transformer for Online 3D Multi-Object Tracking
論文/Paper: http://arxiv.org/pdf/2308.06635
代碼/Code: https://github.com/dsx0511/3dmotformer
DFM-X: Augmentation by Leveraging Prior Knowledge of Shortcut Learning
論文/Paper: http://arxiv.org/pdf/2308.06622
代碼/Code: https://github.com/nis-research/dfmx-augmentation
On the Interplay of Convolutional Padding and Adversarial Robustness
論文/Paper: http://arxiv.org/pdf/2308.06612
代碼/Code: None
Cyclic Test-Time Adaptation on Monocular Video for 3D Human Mesh Reconstruction
論文/Paper: http://arxiv.org/pdf/2308.06554
代碼/Code: https://github.com/hygenie1228/CycleAdapt_RELEASE.
Revisiting Vision Transformer from the View of Path Ensemble
論文/Paper: http://arxiv.org/pdf/2308.06548
代碼/Code: None
SegPrompt: Boosting Open-world Segmentation via Category-level Prompt Learning
論文/Paper: http://arxiv.org/pdf/2308.06531
代碼/Code: https://github.com/aim-uofa/segprompt
BEV-DG: Cross-Modal Learning under Bird's-Eye View for Domain Generalization of 3D Semantic Segmentation
論文/Paper: http://arxiv.org/pdf/2308.06530
代碼/Code: None
Tiny and Efficient Model for the Edge Detection Generalization
論文/Paper: http://arxiv.org/pdf/2308.06468
代碼/Code: https://github.com/xavysp/TEED.
U-RED: Unsupervised 3D Shape Retrieval and Deformation for Partial Point Clouds
論文/Paper: http://arxiv.org/pdf/2308.06383
代碼/Code: https://github.com/zhangcyg/u-red
Unified Data-Free Compression: Pruning and Quantization without Fine-Tuning
論文/Paper: http://arxiv.org/pdf/2308.07209
代碼/Code: None
CBA: Improving Online Continual Learning via Continual Bias Adaptor
論文/Paper: http://arxiv.org/pdf/2308.06925
代碼/Code: https://github.com/wqza/cba-online-cl
Unsupervised Image Denoising in Real-World Scenarios via Self-Collaboration Parallel Generative Adversarial Branches
論文/Paper: http://arxiv.org/pdf/2308.06776
代碼/Code: https://github.com/linxin0/scpgabnet
Multi-Label Knowledge Distillation
論文/Paper: http://arxiv.org/pdf/2308.06453
代碼/Code: https://github.com/penghui-yang/l2d