2023-12-05 Research Report

【1】Disentangled Interaction Representation for One-Stage Human-Object Interaction Detection [arxiv]

标题：用于单级人-物交互检测的分离交互表示法

【2】Multi-task Image Restoration Guided By Robust DINO Features [arxiv]

标题：以鲁棒的 DINO 功能为指导的多任务图像复原

【3】Hyperspectral Image Compression Using Sampling and Implicit Neural Representations [arxiv]

标题：利用采样和隐式神经表征的高光谱图像压缩

【4】VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams [arxiv, code]

标题：将动态辐射场渲染为二维特征视频流

【5】S2P3: Self-Supervised Polarimetric Pose Prediction [IJCV]

标题：自监督极坐标姿态预测

【6】Motion-Guided Latent Diffusion for Temporally Consistent Real-world Video Super-resolution [arxiv]

标题：运动引导潜像扩散，实现时空一致的真实世界视频超分辨率