cs.CV（2024-02-05）

📊 共 24 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (9 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗4) 支柱九：具身大模型 (Embodied Foundation Models) (5 🔗1) 支柱一：机器人控制 (Robot Control) (2) 支柱八：物理动画 (Physics-based Animation) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱三：空间感知与语义 (Perception & Semantics) (9 篇)

#题目一句话要点标签🔗⭐
1 SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM 提出SGS-SLAM以解决神经隐式SLAM的过平滑问题 visual SLAM gaussian splatting splatting
2 4D-Rotor Gaussian Splatting: Towards Efficient Novel View Synthesis for Dynamic Scenes 提出4D-Rotor Gaussian Splatting以解决动态场景的新视角合成问题 3D gaussian splatting gaussian splatting splatting
3 AnaMoDiff: 2D Analogical Motion Diffusion via Disentangled Denoising 提出AnaMoDiff以解决2D运动类比问题 optical flow motion diffusion
4 Denoising Diffusion via Image-Based Rendering 提出一种新型扩散模型以解决3D场景生成问题 3D reconstruction neural radiance field
5 CLIP Can Understand Depth 提出镜像嵌入以改进CLIP在单目深度估计中的表现 depth estimation monocular depth
6 ViewFusion: Learning Composable Diffusion Models for Novel View Synthesis 提出ViewFusion以解决新视角合成中的灵活性问题 NeRF neural radiance field ✅
7 Taylor Videos for Action Recognition 提出Taylor视频以解决动作识别中的运动提取问题 optical flow
8 Motion-Aware Video Frame Interpolation 提出运动感知视频帧插值网络以解决模糊和伪影问题 optical flow
9 Using Motion Cues to Supervise Single-Frame Body Pose and Shape Estimation in Low Data Regimes 利用运动线索监督单帧人体姿态与形状估计以应对数据不足问题 optical flow

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
10 Delving into Multi-modal Multi-task Foundation Models for Road Scene Understanding: From Learning Paradigm Perspectives 提出多模态多任务基础模型以解决道路场景理解问题 world model world models scene understanding ✅
11 FROSTER: Frozen CLIP Is A Strong Teacher for Open-Vocabulary Action Recognition 提出FROSTER以解决开放词汇动作识别问题 distillation open-vocabulary open vocabulary ✅
12 nnMamba: 3D Biomedical Image Segmentation, Classification and Landmark Detection with State Space Model 提出nnMamba以解决3D生物医学图像分割与分类问题 Mamba SSM state space model ✅
13 Retrieval-Augmented Score Distillation for Text-to-3D Generation 提出基于检索增强的评分蒸馏方法以解决文本到3D生成中的几何不一致问题 distillation geometric consistency ✅
14 Constrained Multiview Representation for Self-supervised Contrastive Learning 提出基于约束多视图表示的自监督对比学习方法以提升医学图像分割 representation learning contrastive learning
15 Good Teachers Explain: Explanation-Enhanced Knowledge Distillation 提出解释增强的知识蒸馏方法以提升学生模型性能 distillation

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

#题目一句话要点标签🔗⭐
16 VLN-Video: Utilizing Driving Videos for Outdoor Vision-and-Language Navigation 提出VLN-Video以解决户外视觉语言导航中的环境多样性不足问题 VLN
17 InteractiveVideo: User-Centric Controllable Video Generation with Synergistic Multimodal Instructions 提出InteractiveVideo框架以实现用户中心的视频生成 multimodal ✅
18 Time-, Memory- and Parameter-Efficient Visual Adaptation 提出一种高效视觉适应方法以解决模型微调问题 foundation model
19 Enhancing Compositional Generalization via Compositional Feature Alignment 提出组合特征对齐方法以增强组合泛化能力 foundation model
20 Open-Universe Indoor Scene Generation using LLM Program Synthesis and Uncurated Object Databases 提出一种新系统以生成开放宇宙室内场景 large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
21 Physics-Encoded Graph Neural Networks for Deformation Prediction under Contact 提出物理编码图神经网络以解决接触下变形预测问题 manipulation
22 Visual Text Meets Low-level Vision: A Comprehensive Survey on Visual Text Processing 提出全面的视觉文本处理综述以应对文本特有挑战 manipulation

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

#题目一句话要点标签🔗⭐
23 Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization 提出Video-LaVIT以解决视频语言预训练中的时空建模问题 spatiotemporal large language model multimodal

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

#题目一句话要点标签🔗⭐
24 Exploring Federated Self-Supervised Learning for General Purpose Audio Understanding 提出FASSL框架以解决音频理解中的隐私保护问题 feature matching

⬅️ 返回 cs.CV 首页 · 🏠 返回主页