Ep. 239 - Part 1 - June 5, 2024

ArXiv Computer Vision research for Wednesday, June 05, 2024. 00:20: A Self-Supervised Denoising Strategy for Underwater Acoustic Camera Imageries 01:26: Visual-Text Cross Alignment: Refining the Similarity Score in Vision-Language Models 02:40: U-KAN Makes Strong Backbone for Medical Image Segmentation and Generation 04:14: Exploring Data Efficiency in Zero-Shot Learning with Diffusion Models 06:09: P2PFormer: A Primitive-to-polygon Method for Regular Building Contour Extraction from Remote Sensing Images 08:03: Radiomics-guided Multimodal Self-attention Network for Predicting Pathological Complete Response in Breast MRI 09:08: AVFF: Audio-Visual Feature Fusion for Video Deepfake Detection 10:49: Understanding the Impact of Negative Prompts: When and How Do They Take Effect? 12:03: Adversarial Generation of Hierarchical Gaussians for 3D Generative Model 13:45: Event3DGS: Event-based 3D Gaussian Splatting for Fast Egomotion 15:19: DA-Flow: Dual Attention Normalizing Flow for Skeleton-based Video Anomaly Detection 16:56: Sparse Color-Code Net: Real-Time RGB-Based 6D Object Pose Estimation on Edge Devices 18:07: Self-Supervised Skeleton Action Representation Learning: A Benchmark and Beyond 20:01: Enhancing Multimodal Large Language Models with Multi-instance Visual Prompt Generator for Visual Representation Enrichment 21:05: Predicting Genetic Mutation from Whole Slide Images via Biomedical-Linguistic Knowledge Enhanced Multi-label Classification 22:52: A Human-Annotated Video Dataset for Training and Evaluation of 360-Degree Video Summarization Methods 23:53: EdgeSync: Faster Edge-model Updating via Adaptive Continuous Learning for Video Data Drift 25:13: Phy-Diff: Physics-guided Hourglass Diffusion Model for Diffusion MRI Synthesis 26:34: DriVLMe: Enhancing LLM-based Autonomous Driving Agents with Embodied and Social Experiences 27:58: DifAttack++: Query-Efficient Black-Box Adversarial Attack via Hierarchical Disentangled Feature Space in Cross Domain 29:35: Puzzle Pieces Picker: Deciphering Ancient Chinese Characters with Radical Reconstruction 31:02: Instructing Prompt-to-Prompt Generation for Zero-Shot Learning 32:31: Follow-Your-Pose v2: Multiple-Condition Guided Character Image Animation for Stable Pose Control 34:20: Giving each task what it needs -- leveraging structured sparsity for tailored multi-task learning