cs.CV（2024-02-23）

📊 共 10 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗2) 支柱四：生成式动作 (Generative Motion) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1)

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

#	题目	一句话要点	标签	🔗	⭐
1	CI w/o TN: Context Injection without Task Name for Procedure Planning	提出无任务名称的上下文注入方法以解决程序规划问题	contrastive learning large language model foundation model
2	MambaIR: A Simple Baseline for Image Restoration with State-Space Model	提出MambaIR以解决图像恢复中的局部遗忘与通道冗余问题	Mamba state space model	✅
3	State Space Models for Event Cameras	提出状态空间模型以解决事件相机数据处理的频率适应问题	SSM state space model
4	Cohere3D: Exploiting Temporal Coherence for Unsupervised Representation Learning of Vision-based Autonomous Driving	提出Cohere3D以解决自动驾驶中多帧输入的实例表示学习问题	representation learning contrastive learning
5	Unsupervised Domain Adaptation for Brain Vessel Segmentation through Transwarp Contrastive Learning	提出无监督领域适应方法以改善脑血管分割	predictive model contrastive learning
6	Attention-Guided Masked Autoencoders For Learning Image Representations	提出注意力引导的掩码自编码器以提升图像表示学习	masked autoencoder MAE

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
7	Large Multimodal Agents: A Survey	系统评估大规模多模态智能体以应对复杂任务	large language model multimodal	✅
8	Seeing is Believing: Mitigating Hallucination in Large Vision-Language Models via CLIP-Guided Decoding	提出CLIP引导解码方法以减轻大规模视觉语言模型的幻觉问题	visual grounding	✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
9	Seamless Human Motion Composition with Blended Positional Encodings	提出FlowMDM以解决连续人类动作生成问题	motion generation human motion human motion generation

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
10	OpenSUN3D: 1st Workshop Challenge on Open-Vocabulary 3D Scene Understanding	举办OpenSUN3D挑战赛以推动开放词汇3D场景理解研究	scene understanding open-vocabulary open vocabulary	✅

⬅️ 返回 cs.CV 首页 · 🏠 返回主页