Alert button
Picture for Humphrey Shi

Humphrey Shi

Alert button

Social Reward: Evaluating and Enhancing Generative AI through Million-User Feedback from an Online Creative Community

Feb 15, 2024
Arman Isajanyan, Artur Shatveryan, David Kocharyan, Zhangyang Wang, Humphrey Shi

Viaarxiv icon

VASE: Object-Centric Appearance and Shape Manipulation of Real Videos

Jan 04, 2024
Elia Peruzzo, Vidit Goel, Dejia Xu, Xingqian Xu, Yifan Jiang, Zhangyang Wang, Humphrey Shi, Nicu Sebe

Viaarxiv icon

HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models

Dec 25, 2023
Hayk Manukyan, Andranik Sargsyan, Barsegh Atanyan, Zhangyang Wang, Shant Navasardyan, Humphrey Shi

Viaarxiv icon

VCoder: Versatile Vision Encoders for Multimodal Large Language Models

Dec 21, 2023
Jitesh Jain, Jianwei Yang, Humphrey Shi

Viaarxiv icon

Diffusion for Natural Image Matting

Dec 10, 2023
Yihan Hu, Yiheng Lin, Wei Wang, Yao Zhao, Yunchao Wei, Humphrey Shi

Viaarxiv icon

Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Models

Dec 07, 2023
Jiayi Guo, Xingqian Xu, Yifan Pu, Zanlin Ni, Chaofei Wang, Manushree Vasu, Shiji Song, Gao Huang, Humphrey Shi

Viaarxiv icon

HiFi Tuner: High-Fidelity Subject-Driven Fine-Tuning for Diffusion Models

Nov 30, 2023
Zhonghao Wang, Wei Wei, Yang Zhao, Zhisheng Xiao, Mark Hasegawa-Johnson, Humphrey Shi, Tingbo Hou

Viaarxiv icon

Video Instance Matting

Nov 08, 2023
Jiachen Li, Roberto Henschel, Vidit Goel, Marianna Ohanyan, Shant Navasardyan, Humphrey Shi

Viaarxiv icon

Combating Label Noise With A General Surrogate Model For Sample Selection

Oct 16, 2023
Chao Liang, Linchao Zhu, Humphrey Shi, Yi Yang

Viaarxiv icon

Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else

Oct 11, 2023
Hazarapet Tunanyan, Dejia Xu, Shant Navasardyan, Zhangyang Wang, Humphrey Shi

Figure 1 for Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else
Figure 2 for Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else
Figure 3 for Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else
Figure 4 for Multi-Concept T2I-Zero: Tweaking Only The Text Embeddings and Nothing Else
Viaarxiv icon