cs.CV（2024-03-30）

📊 共 14 篇论文 | 🔗 8 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (8 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (3 🔗2) 支柱四：生成式动作 (Generative Motion) (1 🔗1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱九：具身大模型 (Embodied Foundation Models) (1 🔗1)

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

#题目一句话要点标签🔗⭐
1 3DGSR: Implicit Surface Reconstruction with 3D Gaussian Splatting 提出3DGSR以解决高效3D表面重建问题 3D gaussian splatting 3DGS 3D reconstruction ✅
2 Image-to-Image Matching via Foundation Models: A New Perspective for Open-Vocabulary Semantic Segmentation 提出RIM框架以解决开放词汇语义分割中的错误匹配问题 open-vocabulary open vocabulary foundation model
3 The Devil is in the Edges: Monocular Depth Estimation with Edge-aware Consistency Fusion 提出ECFNet以解决单目深度估计中的边缘信息不足问题 depth estimation monocular depth ✅
4 MaGRITTe: Manipulative and Generative 3D Realization from Image, Topview and Text 提出MaGRITTe以解决多模态条件下3D场景生成问题 depth estimation NeRF neural radiance field
5 Do Vision-Language Models Understand Compound Nouns? 提出Compun基准以评估视觉语言模型对复合名词的理解能力 open-vocabulary open vocabulary large language model ✅
6 DiffHuman: Probabilistic Photorealistic 3D Reconstruction of Humans 提出DiffHuman以解决单张RGB图像的人体3D重建问题 3D reconstruction
7 Reusable Architecture Growth for Continual Stereo Matching 提出可重用架构增长框架以解决持续立体匹配问题 depth estimation stereo depth
8 IPoD: Implicit Field Learning with Point Diffusion for Generalizable 3D Object Reconstruction from Single RGB-D Images 提出IPoD以解决单视图RGB-D图像的3D物体重建问题 implicit representation ✅

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

#题目一句话要点标签🔗⭐
9 Design as Desired: Utilizing Visual Question Answering for Multimodal Pre-training 利用视觉问答实现多模态预训练以解决医学特征学习问题 contrastive learning multimodal ✅
10 HSIMamba: Hyperpsectral Imaging Efficient Feature Learning with Bidirectional State Space for Classification 提出HSIMamba以解决高光谱图像分类中的效率与准确性问题 Mamba HSI
11 TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias 提出TTD方法以解决CLIP模型中的单标签偏差问题 distillation ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
12 Towards Variable and Coordinated Holistic Co-Speech Motion Generation 提出ProbTalk以解决3D虚拟形象的生动共语动作生成问题 motion generation ✅

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

#题目一句话要点标签🔗⭐
13 HOI-M3:Capture Multiple Humans and Objects Interaction within Contextual Environment 提出HOI-M3数据集以解决多人的物体交互建模问题 human-object interaction HOI

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

#题目一句话要点标签🔗⭐
14 ST-LLM: Large Language Models Are Effective Temporal Learners 提出ST-LLM以解决视频理解中的时空建模问题 large language model ✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页