arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

Carnegie Mellon University(卡内基梅隆大学)

2025-12-30 至 2025-12-30 共收录 4
2507.04749 2025-12-30 cs.CV

MatDecompSDF: High-Fidelity 3D Shape and PBR Material Decomposition from Multi-View Images

MatDecompSDF:从多视角图像中恢复高保真3D形状和PBR材质分解

Chengyu Wang, Isabella Bennett, Henry Scott, Liang Zhang, Mei Chen, Hao Li, Rui Zhao

机构 * San Francisco State University 1600 Holloway Avenue San Francisco California USA 94132 Department of Electrical \& Computer Engineering, Boston University 8 Saint Mary’s Street Boston MA USA 02215 University of California, Berkeley 2150 Shattuck Avenue Berkeley California USA 94704 Stanford University 450 Serra Mall Stanford California USA 94305 Carnegie Mellon University 5000 Forbes Avenue Pittsburgh Pennsylvania USA 15213 University of Washington 185 Stevens Way Seattle Washington USA 98195 San Francisco State University Department of Electrical \& Computer Engineering, Boston University University of California, Berkeley Stanford University Carnegie Mellon University University of Washington

AI总结 MatDecompSDF通过联合优化SDF、神经场和MLP模型,实现从多视角图像中高保真3D形状和PBR材质的分解。

Comments 12 pages, 4 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2501.04359 2025-12-30 eess.AS cs.CL cs.HC cs.LG cs.SD

Decoding EEG Speech Perception with Transformers and VAE-based Data Augmentation

基于变换器和基于VAE的数据增强的EEG语音感知解码

Terrance Yu-Hao Chen, Yulin Chen, Pontus Soederhaell, Sadrishya Agrawal, Kateryna Shapovalenko

机构 * Carnegie Mellon University(卡内基梅隆大学)

AI总结 本研究利用VAE和变换器模型提升EEG语音解码性能,通过数据增强和序列到序列架构改进语音感知任务。

Comments 19 pages, 15 figures, 2 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.20981 2025-12-30 cs.CV cs.CL cs.RO

RefAV: Towards Planning-Centric Scenario Mining

RefAV:面向规划导向的场景挖掘

Cainan Davidson, Deva Ramanan, Neehar Peri

机构 * Carnegie Mellon University(卡内基梅隆大学)

AI总结 RefAV通过视觉-语言模型改进场景挖掘,解决自动驾驶中复杂多智能体交互的定位问题。

Comments Project Page: https://cainand.github.io/RefAV/

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.06474 2025-12-30 cs.CV cs.AI cs.LG

Enhancing Vision-Language Model Reliability with Uncertainty-Guided Dropout Decoding

通过不确定性引导的丢弃解码增强视觉-语言模型的可靠性

Yixiong Fang, Ziran Yang, Zhaorun Chen, Zhuokai Zhao, Jiawei Zhou

机构 * Carnegie Mellon University(卡内基梅隆大学) Princeton University(普林斯顿大学) University of Chicago(芝加哥大学) Stony Brook University(石溪大学)

AI总结 通过不确定性引导的丢弃解码方法,有效减少视觉-语言模型的幻觉问题,提升输出的可靠性和质量。

Comments Accepted to 39th Conference on Neural Information Processing Systems (NeurIPS 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏