Výsledky vyhledávání

Report

Interpreting Low-level Vision Models with Causal Effect Maps

Autor: Hu, Jinfan, Gu, Jinjin, Yu, Shiyao, Yu, Fanghua, Li, Zheyuan, You, Zhiyuan, Lu, Chaochao, Dong, Chao

Deep neural networks have significantly improved the performance of low-level vision tasks but also increased the difficulty of interpretability. A deep understanding of deep models is beneficial for both network design and practical reliability. To

Externí odkaz: http://arxiv.org/abs/2407.19789

Zobrazit plný text záznamu

Report

Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild

Autor: Yu, Fanghua, Gu, Jinjin, Li, Zheyuan, Hu, Jinfan, Kong, Xiangtao, Wang, Xintao, He, Jingwen, Qiao, Yu, Dong, Chao

We introduce SUPIR (Scaling-UP Image Restoration), a groundbreaking image restoration method that harnesses generative prior and the power of model scaling up. Leveraging multi-modal techniques and advanced generative prior, SUPIR marks a significant

Externí odkaz: http://arxiv.org/abs/2401.13627

Zobrazit plný text záznamu

Report

DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior

Autor: Lin, Xinqi, He, Jingwen, Chen, Ziyan, Lyu, Zhaoyang, Dai, Bo, Yu, Fanghua, Ouyang, Wanli, Qiao, Yu, Dong, Chao

We present DiffBIR, a general restoration pipeline that could handle different blind image restoration tasks in a unified framework. DiffBIR decouples blind image restoration problem into two stages: 1) degradation removal: removing image-independent

Externí odkaz: http://arxiv.org/abs/2308.15070

Zobrazit plný text záznamu

Report

GET3D--: Learning GET3D from Unconstrained Image Collections

Autor: Yu, Fanghua, Wang, Xintao, Li, Zheyuan, Cao, Yan-Pei, Shan, Ying, Dong, Chao

The demand for efficient 3D model generation techniques has grown exponentially, as manual creation of 3D models is time-consuming and requires specialized expertise. While generative models have shown potential in creating 3D textured shapes from 2D

Externí odkaz: http://arxiv.org/abs/2307.14918

Zobrazit plný text záznamu

Report

OSRT: Omnidirectional Image Super-Resolution with Distortion-aware Transformer

Autor: Yu, Fanghua, Wang, Xintao, Cao, Mingdeng, Li, Gen, Shan, Ying, Dong, Chao

Omnidirectional images (ODIs) have obtained lots of research interest for immersive experiences. Although ODIs require extremely high resolution to capture details of the entire scene, the resolutions of most ODIs are insufficient. Previous methods a

Externí odkaz: http://arxiv.org/abs/2302.03453

Zobrazit plný text záznamu