cs.CV(2024-10-22)

📊 共 7 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱三:空间感知与语义 (Perception & Semantics) (3 🔗2) 支柱八:物理动画 (Physics-based Animation) (1) 支柱九:具身大模型 (Embodied Foundation Models) (1 🔗1) 支柱二:RL算法与架构 (RL & Architecture) (1) 支柱一:机器人控制 (Robot Control) (1 🔗1)

🔬 支柱三:空间感知与语义 (Perception & Semantics) (3 篇)

#题目一句话要点标签🔗
1 E-3DGS: Gaussian Splatting with Exposure and Motion Events E-3DGS:利用曝光和运动事件进行高鲁棒性、低成本的3D高斯溅射重建 3D gaussian splatting 3DGS gaussian splatting
2 SpectroMotion: Dynamic 3D Reconstruction of Specular Scenes SpectroMotion:结合3DGS与PBR,实现动态高光场景的3D重建 3D gaussian splatting 3DGS gaussian splatting
3 LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias LVSM:一种基于Transformer的极少3D先验知识的大规模视角合成模型 3DGS NeRF

🔬 支柱八:物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗
4 LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding LongVU:时空自适应压缩长视频,提升视频语言理解能力 spatiotemporal large language model multimodal

🔬 支柱九:具身大模型 (Embodied Foundation Models) (1 篇)

#题目一句话要点标签🔗
5 VideoSAM: A Large Vision Foundation Model for High-Speed Video Segmentation VideoSAM:用于高速视频分割的大型视觉基础模型,提升复杂相检测精度。 foundation model

🔬 支柱二:RL算法与架构 (RL & Architecture) (1 篇)

#题目一句话要点标签🔗
6 SigCLR: Sigmoid Contrastive Learning of Visual Representations SigCLR:提出基于Sigmoid函数的对比学习视觉表征方法 contrastive learning

🔬 支柱一:机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗
7 Denoise-I2W: Mapping Images to Denoising Words for Accurate Zero-Shot Composed Image Retrieval 提出Denoise-I2W,通过图像到去噪词映射提升零样本组合图像检索精度 manipulation

⬅️ 返回 cs.CV 首页 · 🏠 返回主页