Computer Vision – ECCV 2024

Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptation.- Learning to Obstruct Few-Shot Image Classification over Restricted Classes.- RoofDiffusion:  Constructing Roofs from Severely Corrupted Point Data via Diffusion.- L-DiffER: Single Image Reflection Removal with Language-based Diffusion Model.- AdaShield: Safeguarding Multimodal Large Language Models from Structure-based Attack via Adaptive Shield Prompting.- OccGen: Generative Multi-modal 3D Occupancy Prediction for Autonomous Driving.- CrossGLG: LLM Guides One-shot Skeleton-based 3D Action Recognition in a Cross-level Manner.- HYDRA: A Hyper Agent for Dynamic Compositional Visual Reasoning.- BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion.- LayoutDETR: Detection Transformer Is a Good Multimodal Layout Designer.- Blind image deblurring with noise-robust kernel estimation.- Binomial Self-compensation for Motion Error in Dynamic 3D Scanning.- AddMe: Zero-shot Group-photo Synthesis by Inserting People into Scenes.- Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation.- VersatileGaussian: Real-time Neural Rendering for Versatile Tasks using Gaussian Splatting.- Momentum Auxiliary Network for Supervised Local Learning.- HPFF: Hierarchical Locally Supervised Learning with Patch Feature Fusion.- Rethinking LiDAR Domain Generalization: Single Source as Multiple Density Domains.- Improving Zero-Shot Generalization for CLIP with Variational Adapter.- Realistic Human Motion Generation with Cross-Diffusion Models.- EgoExo-Fitness: Towards Egocentric and Exocentric Full-Body Action Understanding.- Any Target Can be Offense: Adversarial Example Generation via Generalized Latent Infection.- Towards Reliable Advertising Image Generation Using Human Feedback.- Topology-Preserving Downsampling of Binary Images.- ColorMAE: Exploring data-independent masking strategies in Masked AutoEncoders.- Classification Matters: Improving Video Action Detection with Class-Specific Attention.- Improving Medical Multi-modal Contrastive Learning with Expert Annotations.