cs.CV（2024-02-27）

📊 共 27 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (8 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (8) 支柱九：具身大模型 (Embodied Foundation Models) (5 🔗2) 支柱一：机器人控制 (Robot Control) (2) 支柱六：视频提取与匹配 (Video Extraction) (1 🔗1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
1 Structural Teacher-Student Normality Learning for Multi-Class Anomaly Detection and Localization 提出结构化教师-学生正常性学习以解决多类异常检测问题 teacher-student distillation
2 ShapeLLM: Universal 3D Object Understanding for Embodied Interaction 提出ShapeLLM以解决3D对象理解与交互问题 distillation large language model multimodal ✅
3 MCF-VC: Mitigate Catastrophic Forgetting in Class-Incremental Learning for Multimodal Video Captioning 提出MCF-VC以解决多模态视频字幕中的灾难性遗忘问题 distillation multimodal
4 LiveHPS: LiDAR-based Scene-level Human Pose and Shape Estimation in Free Environment 提出LiveHPS以解决自由环境下的人体姿态与形状估计问题 distillation scene understanding human motion
5 Towards Robust and Efficient Cloud-Edge Elastic Model Adaptation via Selective Entropy Distillation 提出云边弹性模型适应方法以应对动态环境挑战 distillation foundation model
6 Demonstrating and Reducing Shortcuts in Vision-Language Representation Learning 提出合成快捷方式以解决视觉-语言表示学习中的学习偏差问题 representation learning contrastive learning
7 PLReMix: Combating Noisy Labels with Pseudo-Label Relaxed Contrastive Representation Learning 提出PLReMix以解决噪声标签下的对比学习问题 representation learning ✅
8 LSPT: Long-term Spatial Prompt Tuning for Visual Representation Learning 提出长时空提示调优方法以解决视觉表示学习中的遗忘问题 representation learning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (8 篇)

#题目一句话要点标签🔗⭐
9 VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction 提出VastGaussian以解决大场景重建中的视觉质量与渲染速度问题 3D gaussian splatting gaussian splatting splatting
10 Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis 提出动态四面体模型以解决高质量人头合成问题 NeRF neural radiance field implicit representation
11 AVS-Net: Point Sampling with Adaptive Voxel Size for 3D Scene Understanding 提出AVS-Net以解决大规模3D场景下的点云下采样问题 scene understanding
12 ICP-Flow: LiDAR Scene Flow Estimation with ICP 提出ICP-Flow以解决LiDAR场景流估计问题 scene flow
13 CharNeRF: 3D Character Generation from Concept Art 提出CharNeRF以解决3D角色生成的效率问题 3D reconstruction NeRF neural radiance field
14 Sora Generates Videos with Stunning Geometrical Consistency 提出新基准评估Sora模型生成视频的几何一致性 3D reconstruction
15 A Vanilla Multi-Task Framework for Dense Visual Prediction Solution to 1st VCL Challenge -- Multi-Task Robustness Track 提出UniNet框架以解决视觉持续学习中的多任务鲁棒性问题 depth estimation
16 DivAvatar: Diverse 3D Avatar Generation with a Single Prompt 提出DivAvatar以解决3D头像生成多样性不足问题 NeRF

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#题目一句话要点标签🔗⭐
17 Multimodal Learned Sparse Retrieval with Probabilistic Expansion Control 提出多模态学习稀疏检索方法以解决现有模型复杂性问题 multimodal
18 AlignMiF: Geometry-Aligned Multimodal Implicit Field for LiDAR-Camera Joint Synthesis 提出AlignMiF以解决LiDAR与相机数据融合中的几何对齐问题 multimodal
19 Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology 提出R$^2$T以解决计算病理学中MIL模型适应性不足问题 foundation model ✅
20 Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling 提出Masked Gamma-SSL以解决不确定性估计问题 foundation model
21 Seeing and Hearing: Open-domain Visual-Audio Generation with Diffusion Latent Aligners 提出多模态潜在对齐器以解决视频音频生成问题 multimodal ✅

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
22 Differentiable Biomechanics Unlocks Opportunities for Markerless Motion Capture 提出可微生物力学方法以解决无标记运动捕捉问题 trajectory optimization implicit representation markerless motion capture
23 One-Shot Structure-Aware Stylized Image Synthesis 提出OSASIS以解决图像风格化中的结构保持问题 manipulation structure preservation

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
24 OSCaR: Object State Captioning and State Change Representation 提出OSCaR数据集以解决对象状态变化理解问题 egocentric large language model multimodal ✅

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

#题目一句话要点标签🔗⭐
25 V2C-Long: Longitudinal Cortex Reconstruction with Spatiotemporal Correspondence 提出V2C-Long以解决纵向MRI皮层重建中的时空对应问题 spatiotemporal

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
26 SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents 提出SocialCVAE以解决行人轨迹预测中的不确定性问题 human motion

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
27 ADL4D: Towards A Contextually Rich Dataset for 4D Activities of Daily Living 提出ADL4D数据集以解决4D日常活动中的上下文缺失问题 HOI HMR

⬅️ 返回 cs.CV 首页 · 🏠 返回主页