cs.CV（2025-04-13）

📊 共 12 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (4 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (3 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱七：动作重定向 (Motion Retargeting) (1 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
1 SegEarth-R1: Geospatial Pixel Reasoning via Large Language Model 提出SegEarth-R1，通过大语言模型实现地理空间像素推理，解决遥感图像的复杂查询问题。 large language model ✅
2 Ges3ViG: Incorporating Pointing Gestures into Language-Based 3D Visual Grounding for Embodied Reference Understanding Ges3ViG：融合指向手势的语言3D视觉定位，提升具身引用理解 visual grounding ✅
3 Automatic Detection of Intro and Credits in Video using CLIP and Multihead Attention 提出基于CLIP和多头注意力机制的视频片头片尾自动检测方法，提升内容理解效率。 multimodal
4 Low-Light Image Enhancement using Event-Based Illumination Estimation RetinEV：利用事件相机的时间映射事件进行低光照图像增强 TAMP

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

#题目一句话要点标签🔗⭐
5 TextSplat: Text-Guided Semantic Fusion for Generalizable Gaussian Splatting TextSplat：文本引导的语义融合，提升可泛化高斯溅射重建效果 gaussian splatting splatting geometric consistency
6 Vision-Language Model for Object Detection and Segmentation: A Review and Evaluation 系统性评测视觉-语言模型在目标检测与分割任务中的性能与局限性 open-vocabulary open vocabulary foundation model ✅
7 DropoutGS: Dropping Out Gaussians for Better Sparse-view Rendering DropoutGS：通过高斯随机丢弃提升稀疏视角下的3DGS渲染效果 3D gaussian splatting 3DGS gaussian splatting ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
8 Sparse Deformable Mamba for Hyperspectral Image Classification 提出稀疏可变形Mamba模型，提升高光谱图像分类精度与效率。 Mamba HSI
9 ERL-MPP: Evolutionary Reinforcement Learning with Multi-head Puzzle Perception for Solving Large-scale Jigsaw Puzzles of Eroded Gaps 提出ERL-MPP框架，解决带侵蚀间隙的大规模拼图难题。 reinforcement learning
10 FSSUAVL: A Discriminative Framework using Vision Models for Federated Self-Supervised Audio and Image Understanding 提出FSSUAVL，利用联邦自监督学习解决非配对音频图像理解问题 contrastive learning multimodal

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

#题目一句话要点标签🔗⭐
11 EmbodiedOcc++: Boosting Embodied 3D Occupancy Prediction with Plane Regularization and Uncertainty Sampler 提出EmbodiedOcc++以解决室内三维占用预测中的几何特征不足问题 geometric consistency ✅

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

#题目一句话要点标签🔗⭐
12 FVOS for MOSE Track of 4th PVUW Challenge: 3rd Place Solution 提出FVOS以解决复杂场景下视频目标分割问题 spatiotemporal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页