cs.CV(2024-05-29)

📊 共 27 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱九:具身大模型 (Embodied Foundation Models) (10 🔗4) 支柱三:空间感知与语义 (Perception & Semantics) (4) 支柱二:RL算法与架构 (RL & Architecture) (4 🔗2) 支柱一:机器人控制 (Robot Control) (3 🔗1) 支柱六:视频提取与匹配 (Video Extraction) (3 🔗2) 支柱七:动作重定向 (Motion Retargeting) (1) 支柱四:生成式动作 (Generative Motion) (1) 支柱八:物理动画 (Physics-based Animation) (1)

🔬 支柱九:具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗
1 X-VILA: Cross-Modality Alignment for Large Language Model X-VILA:通过跨模态对齐扩展大型语言模型至图像、视频和音频 large language model instruction following
2 Kestrel: 3D Multimodal LLM for Part-Aware Grounded Description Kestrel:提出一种3D多模态LLM,用于零件感知的三维场景描述与定位。 large language model multimodal
3 Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare 提出Compare2Score,通过教导大模型比较图像质量,实现自适应图像质量评估。 multimodal
4 Multi-Modal Generative Embedding Model 提出多模态生成式嵌入模型MM-GEM,统一生成与嵌入任务,实现细粒度多模态理解。 large language model multimodal
5 Cephalo: Multi-Modal Vision-Language Models for Bio-Inspired Materials Analysis and Design Cephalo:用于生物启发材料分析与设计的视觉-语言多模态模型 large language model multimodal
6 Correctable Landmark Discovery via Large Models for Vision-Language Navigation 提出CONSOLE框架,利用大模型进行可纠正的地标发现,提升视觉-语言导航性能 VLN
7 LLM-based Hierarchical Concept Decomposition for Interpretable Fine-Grained Image Classification 提出基于LLM的分层概念分解Hi-CoDe框架,用于可解释的细粒度图像分类 large language model
8 Enhancing Zero-Shot Facial Expression Recognition by LLM Knowledge Transfer 提出Exp-CLIP,利用LLM知识迁移增强零样本面部表情识别 large language model
9 Benchmarking and Improving Detail Image Caption 提出CAPTURE基准和评估指标,提升大型视觉语言模型在细节图像描述任务上的性能 multimodal
10 WTTFNet: A Weather-Time-Trajectory Fusion Network for Pedestrian Trajectory Prediction in Urban Complex WTTFNet:融合天气-时间-轨迹信息的行人轨迹预测网络,提升城市复杂环境下的预测精度。 multimodal

🔬 支柱三:空间感知与语义 (Perception & Semantics) (4 篇)

#题目一句话要点标签🔗
11 Parameter-efficient Fine-tuning in Hyperspherical Space for Open-vocabulary Semantic Segmentation 提出H-CLIP,一种在超球面空间进行参数高效微调的开放词汇语义分割方法 open-vocabulary open vocabulary foundation model
12 NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild NeRF On-the-go:利用不确定性消除干扰,实现复杂场景下鲁棒的NeRF重建 NeRF neural radiance field
13 Neural Radiance Fields for Novel View Synthesis in Monocular Gastroscopy 提出基于几何先验的神经辐射场方法,用于单目胃镜图像的新视角合成 NeRF neural radiance field
14 NPGA: Neural Parametric Gaussian Avatars 提出神经参数高斯头像(NPGA),用于高保真、可控人头数字重建与实时渲染。 3D gaussian splatting gaussian splatting splatting

🔬 支柱二:RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗
15 Vim-F: Visual State Space Model Benefiting from Learning in the Frequency Domain 提出Vim-F,利用频域学习增强视觉状态空间模型,提升图像建模能力。 Mamba SSM state space model
16 FourierMamba: Fourier Learning Integration with State Space Models for Image Deraining 提出FourierMamba,利用傅里叶空间的状态空间模型进行图像去雨 Mamba state space model
17 LetsMap: Unsupervised Representation Learning for Semantic BEV Mapping 提出LetsMap,一种无监督语义BEV地图表示学习方法,解决自动驾驶场景中标注数据依赖问题。 representation learning masked autoencoder
18 DGD: Dynamic 3D Gaussians Distillation 提出DGD,通过动态3D高斯蒸馏实现单目视频动态语义辐射场学习。 distillation

🔬 支柱一:机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗
19 Reasoning3D -- Grounding and Reasoning in 3D: Fine-Grained Zero-Shot Open-Vocabulary 3D Reasoning Part Segmentation via Large Vision-Language Models 提出 Reasoning3D,通过大型视觉-语言模型实现细粒度零样本开放词汇3D推理分割。 manipulation open-vocabulary open vocabulary
20 Diffusion Policy Attacker: Crafting Adversarial Attacks for Diffusion-based Policies 提出DP-Attacker,针对扩散策略的离线、在线、全局和局部对抗攻击方法 manipulation behavior cloning diffusion policy
21 Evaluating Vision-Language Models on Bistable Images 利用双稳态图像评估视觉-语言模型对歧义感知的鲁棒性 manipulation

🔬 支柱六:视频提取与匹配 (Video Extraction) (3 篇)

#题目一句话要点标签🔗
22 MEGA: Masked Generative Autoencoder for Human Mesh Recovery 提出MEGA:用于人体网格恢复的掩码生成自编码器,实现确定性和随机性模式下的SOTA性能。 human mesh recovery HMR
23 Single image super-resolution based on trainable feature matching attention network 提出可训练特征匹配注意力网络TFMAN,用于提升单图像超分辨率重建性能。 feature matching
24 $E^{3}$Gen: Efficient, Expressive and Editable Avatars Generation 提出E³Gen,实现高效、富有表现力且可编辑的3D头像生成 SMPL SMPL-X

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

#题目一句话要点标签🔗
25 Multi-Condition Latent Diffusion Network for Scene-Aware Neural Human Motion Prediction 提出多条件潜在扩散网络MCLD,用于场景感知的神经人体运动预测。 human motion human motion prediction motion prediction

🔬 支柱四:生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗
26 Programmable Motion Generation for Open-Set Motion Control Tasks 提出可编程运动生成框架,解决开放场景下可定制的角色动画控制问题 motion generation character animation large language model

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

#题目一句话要点标签🔗
27 SFANet: Spatial-Frequency Attention Network for Weather Forecasting 提出SFANet,利用空间-频率注意力机制提升天气预报精度。 spatiotemporal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页