Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report

• URL: http://arxiv.org/abs/2211.05910v1
• Date: Mon, 7 Nov 2022 22:27:58 GMT
• Title: Efficient and Accurate Quantized Image Super-Resolution on Mobile NPUs, Mobile AI & AIM 2022 challenge: Report
• Authors: Andrey Ignatov and Radu Timofte and Maurizio Denna and Abdel Younes and Ganzorig Gankhuyag and Jingang Huh and Myeong Kyun Kim and Kihwan Yoon and Hyeon-Cheol Moon and Seungho Lee and Yoonsik Choe and Jinwoo Jeong and Sungjei Kim and Maciej Smyl and Tomasz Latkowski and Pawel Kubik and Michal Sokolski and Yujie Ma and Jiahao Chao and Zhou Zhou and Hongfan Gao and Zhengfeng Yang and Zhenbing Zeng and Zhengyang Zhuge and Chenghua Li and Dan Zhu and Mengdi Sun and Ran Duan and Yan Gao and Lingshun Kong and Long Sun and Xiang Li and Xingdong Zhang and Jiawei Zhang and Yaqi Wu and Jinshan Pan and Gaocheng Yu and Jin Zhang and Feng Zhang and Zhe Ma and Hongbin Wang and Hojin Cho and Steve Kim and Huaen Li and Yanbo Ma and Ziwei Luo and Youwei Li and Lei Yu and Zhihong Wen and Qi Wu and Haoqiang Fan and Shuaicheng Liu and Lize Zhang and Zhikai Zong and Jeremy Kwon and Junxi Zhang and Mengyuan Li and Nianxiang Fu and Guanchen Ding and Han Zhu and Zhenzhong Chen and Gen Li and Yuanfan Zhang and Lei Sun and Dafeng Zhang and Neo Yang and Fitz Liu and Jerry Zhao and Mustafa Ayazoglu and Bahri Batuhan Bilecen and Shota Hirose and Kasidis Arunruangsirilert and Luo Ao and Ho Chun Leung and Andrew Wei and Jie Liu and Qiang Liu and Dahai Yu and Ao Li and Lei Luo and Ce Zhu and Seongmin Hong and Dongwon Park and Joonhee Lee and Byeong Hyun Lee and Seunggyu Lee and Se Young Chun and Ruiyuan He and Xuhao Jiang and Haihang Ruan and Xinjian Zhang and Jing Liu and Garas Gendy and Nabil Sabor and Jingchao Hou and Guanghui He
• Abstract summary: In this paper, we propose the participants to design an efficient quantized image super-resolution solution. The solution can demonstrate a real-time performance on mobile NPUs. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU.
• Score: 144.41960648643564
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper.

Related papers

• Power Efficient Video Super-Resolution on Mobile NPUs with Deep Learning, Mobile AI & AIM 2022 challenge: Report [97.01510729548531]
  We propose a real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. Models were evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption.
  arXiv  Detail & Related papers  (2022-11-07T22:33:19Z)
• Real-Time Video Super-Resolution on Smartphones with Deep Learning, Mobile AI 2021 Challenge: Report [135.69469815238193]
  Video super-resolution has become one of the most important mobile-related problems due to the rise of video communication and streaming services. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based video super-resolution solutions. The proposed solutions are fully compatible with any mobile GPU and can upscale videos to HD resolution at up to 80 FPS while demonstrating high fidelity results.
  arXiv  Detail & Related papers  (2021-05-17T13:40:50Z)
• Real-Time Quantized Image Super-Resolution on Mobile NPUs, Mobile AI 2021 Challenge: Report [67.86837649834636]
  We introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based image super-resolution solution. The proposed solutions are fully compatible with all major mobile AI accelerators and are capable of reconstructing Full HD images under 40-60 ms.
  arXiv  Detail & Related papers  (2021-05-17T13:34:15Z)
• Fast Camera Image Denoising on Mobile GPUs with Deep Learning, Mobile AI 2021 Challenge: Report [64.09439666916465]
  We introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based image denoising solution. The proposed solutions are fully compatible with any mobile GPU and are capable of processing 480p resolution images under 40-80 ms.
  arXiv  Detail & Related papers  (2021-05-17T13:27:56Z)
• Learned Smartphone ISP on Mobile NPUs with Deep Learning, Mobile AI 2021 Challenge: Report [49.643297263102845]
  This challenge aims to develop an end-to-end deep learning-based image signal processing pipeline. The proposed solutions are capable of processing Full HD photos under 60-100 milliseconds while achieving high fidelity results.
  arXiv  Detail & Related papers  (2021-05-17T13:20:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.