cs.CV（2024-03-11）

📊 共 32 篇论文 | 🔗 10 篇有代码

🎯 兴趣领域导航

支柱三：空间感知与语义 (Perception & Semantics) (9 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (6 🔗2) 支柱一：机器人控制 (Robot Control) (4 🔗1) 支柱七：动作重定向 (Motion Retargeting) (3 🔗1) 支柱八：物理动画 (Physics-based Animation) (1 🔗1) 支柱四：生成式动作 (Generative Motion) (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 FreGS: 3D Gaussian Splatting with Progressive Frequency Regularization 提出FreGS以解决3D高斯重构模糊问题 3D gaussian splatting gaussian splatting splatting
2 Real-time Transformer-based Open-Vocabulary Detection with Efficient Fusion Head 提出OmDet-Turbo以解决实时开放词汇检测问题 open-vocabulary open vocabulary ✅
3 DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization 提出DNGaussian以解决稀疏视图下3D高斯辐射场的训练与推理效率问题 monocular depth 3D gaussian splatting gaussian splatting
4 STARFlow: Spatial Temporal Feature Re-embedding with Attentive Learning for Real-world Scene Flow 提出STARFlow以解决动态场景流预测中的长依赖匹配问题 scene flow spatiotemporal ✅
5 Transferring Relative Monocular Depth to Surgical Vision with Temporal Consistency 提出时间一致性自监督方法以提升外科视觉中的相对单目深度传递 monocular depth
6 Forest Inspection Dataset for Aerial Semantic Segmentation and Depth Estimation 提出森林检查数据集以解决森林监测中的数据不足问题 depth estimation
7 LISO: Lidar-only Self-Supervised 3D Object Detection 提出LISO以解决激光雷达自监督3D目标检测问题 optical flow scene flow
8 Voxel-Mesh Hybrid Representation for Real-Time View Synthesis 提出Vosh以解决实时视图合成中的渲染质量与速度问题 NeRF neural radiance field ✅
9 Bayesian Diffusion Models for 3D Shape Reconstruction 提出贝叶斯扩散模型以解决3D形状重建问题 3D reconstruction

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#题目一句话要点标签🔗⭐
10 DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation 提出DriveDreamer-2以解决定制化驾驶视频生成问题 world model world models dreamer
11 VideoMamba: State Space Model for Efficient Video Understanding 提出VideoMamba以解决视频理解中的冗余与依赖问题 Mamba state space model distillation ✅
12 Large Model driven Radiology Report Generation with Clinical Quality Reinforcement Learning 提出LM-RRG以解决放射科报告生成的临床质量问题 reinforcement learning large language model multimodal
13 Point Mamba: A Novel Point Cloud Backbone Based on State Space Model with Octree-Based Ordering Strategy 提出Point Mamba以解决点云处理中的因果性问题 Mamba SSM state space model ✅
14 One Category One Prompt: Dataset Distillation using Diffusion Models 提出基于扩散模型的数据集蒸馏方法以解决存储挑战 distillation foundation model
15 LeOCLR: Leveraging Original Images for Contrastive Learning of Visual Representations 提出LeOCLR以解决对比学习中的语义特征损失问题 representation learning contrastive learning
16 Enhancing Image Caption Generation Using Reinforcement Learning with Human Feedback 提出基于人类反馈的强化学习方法以提升图像描述生成 reinforcement learning RLHF

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#题目一句话要点标签🔗⭐
17 Can LLMs' Tuning Methods Work in Medical Multimodal Domain? 提出参数高效微调方法以提升医疗多模态模型性能 large language model multimodal ✅
18 Human Pose Descriptions and Subject-Focused Attention for Improved Zero-Shot Transfer in Human-Centric Classification Tasks 提出基于LLM的姿态描述以提升人类中心分类任务的零-shot转移能力 zero-shot transfer
19 Leveraging Foundation Models for Content-Based Image Retrieval in Radiology 利用基础模型提升放射学中的基于内容的图像检索 foundation model ✅
20 PointSeg: A Training-Free Paradigm for 3D Scene Segmentation via Foundation Models 提出PointSeg以解决3D场景分割问题 foundation model
21 A Segmentation Foundation Model for Diverse-type Tumors 提出肿瘤分割基础模型以解决医学图像数据不足问题 foundation model
22 DiaLoc: An Iterative Approach to Embodied Dialog Localization 提出DiaLoc以解决对话基础定位任务的不足问题 multimodal

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#题目一句话要点标签🔗⭐
23 Put Myself in Your Shoes: Lifting the Egocentric Perspective from Exocentric Videos 提出Exo2Ego以解决第三人称视频到第一人称视角转换问题 manipulation egocentric
24 Real-Time Simulated Avatar from Head-Mounted Sensors 提出SimXR以解决头戴传感器控制虚拟化身的问题 humanoid humanoid control egocentric
25 Genetic Learning for Designing Sim-to-Real Data Augmentations 提出遗传学习方法以自动设计数据增强策略解决仿真与现实间差距问题 sim-to-real
26 Fast Text-to-3D-Aware Face Generation and Manipulation via Direct Cross-modal Mapping and Geometric Regularization 提出E^3-FaceNet以解决低效低质的3D人脸生成问题 manipulation ✅

🔬 支柱七：动作重定向 (Motion Retargeting) (3 篇)

#题目一句话要点标签🔗⭐
27 DivCon: Divide and Conquer for Complex Numerical and Spatial Reasoning in Text-to-Image Generation 提出分治策略以解决复杂文本到图像生成中的数值与空间推理问题 spatial relationship large language model
28 Style2Talker: High-Resolution Talking Head Generation with Emotion Style and Art Style 提出Style2Talker以解决音频驱动的表情与艺术风格生成问题 motion representation
29 BEV$^2$PR: BEV-Enhanced Visual Place Recognition with Structural Cues 提出BEV$^2$PR以解决图像基础视觉位置识别中的结构信息不足问题 spatial relationship ✅

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

#题目一句话要点标签🔗⭐
30 Attention Prompt Tuning: Parameter-efficient Adaptation of Pre-trained Models for Spatiotemporal Modeling 提出注意力提示调优以解决视频建模中的参数效率问题 spatiotemporal ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
31 Say Anything with Any Style 提出SAAS方法以解决自然视频生成中的风格多样性问题 VQ-VAE

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

#题目一句话要点标签🔗⭐
32 Ensemble Quadratic Assignment Network for Graph Matching 提出集成二次分配网络以解决图匹配问题 feature matching

⬅️ 返回 cs.CV 首页 · 🏠 返回主页