Výsledky vyhledávání

Towards Accurate Post-Training Quantization for Vision Transformer

Autor: Yifu Ding, Haotong Qin, Qinghua Yan, Zhenhua Chai, Junjie Liu, Xiaolin Wei, Xianglong Liu

Vision transformer emerges as a potential architecture for vision tasks. However, the intense computation and non-negligible delay hinder its application in the real world. As a widespread model compression technique, existing post-training quantizat

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f1a8aee9233d2379be724e48edd9a254
http://arxiv.org/abs/2303.14341

Zobrazit plný text záznamu

An Empirical study of Data-Free Quantization’s Tuning Robustness

Autor: Hong Chen, Yuxuan Wen, Yifu Ding, Zhen Yang, Yufei Guo, Haotong Qin

Publikováno v: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::d292744da82ecdcf31e86cb81e424dbd
https://doi.org/10.1109/cvprw56347.2022.00029

Zobrazit plný text záznamu

BiFSMNv2: Pushing Binary Neural Networks for Keyword Spotting to Real-Network Performance

Autor: Haotong Qin, Xudong Ma, Yifu Ding, Xiaoyang Li, Yang Zhang, Zejun Ma, Jiakai Wang, Jie Luo, Xianglong Liu

Deep neural networks, such as the Deep-FSMN, have been widely studied for keyword spotting (KWS) applications while suffering expensive computation and storage. Therefore, network compression technologies like binarization are studied to deploy KWS m

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::166ea3826e73fb4fa57ddc8e206a4181

Zobrazit plný text záznamu

BiFSMN: Binary Neural Network for Keyword Spotting

Autor: Haotong Qin, Xudong Ma, Yifu Ding, Xiaoyang Li, Yang Zhang, Yao Tian, Zejun Ma, Jie Luo, Xianglong Liu

The deep neural networks, such as the Deep-FSMN, have been widely studied for keyword spotting (KWS) applications. However, computational resources for these networks are significantly constrained since they usually run on-call on edge devices. In th

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1109d0d00ec7977f37ff72a4ef218d03

Zobrazit plný text záznamu

Improving Generalization of Deepfake Detection with Domain Adaptive Batch Normalization

Autor: Zixin Yin, Jun Guo, Yisong Xiao, Haotong Qin, Jiakai Wang, Yifu Ding, Renshuai Tao

Publikováno v: AdvM @ ACM Multimedia

Deepfake, a well-known face forgery technique, has raised serious concerns about personal privacy and social media security. Therefore, a plenty of deepfake detection methods come out and achieve outstanding performance in the single dataset case. Ho

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::51aee4d7dbb9edf8be153292749d700b
https://doi.org/10.1145/3475724.3483603

Zobrazit plný text záznamu

Distribution-sensitive Information Retention for Accurate Binary Neural Network

Autor: Haotong Qin, Xiangguo Zhang, Ruihao Gong, Yifu Ding, Yi Xu, Xianglong Liu

Model binarization is an effective method of compressing neural networks and accelerating their inference process. However, a significant performance gap still exists between the 1-bit model and the 32-bit one. The empirical study shows that binariza

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::34387c51a93701a4ab98a334e4076597
http://arxiv.org/abs/2109.12338

Zobrazit plný text záznamu

Diverse Sample Generation: Pushing the Limit of Generative Data-free Quantization

Autor: Haotong Qin, Yifu Ding, Xiangguo Zhang, Jiakai Wang, Xianglong Liu, Jiwen Lu

Generative data-free quantization emerges as a practical compression approach that quantizes deep neural networks to low bit-width without accessing the real data. This approach generates data utilizing batch normalization (BN) statistics of the full

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::06b55331d4ad1666edb31fe18daa14e9
http://arxiv.org/abs/2109.00212

Zobrazit plný text záznamu

Diversifying Sample Generation for Accurate Data-Free Quantization

Autor: Yifu Ding, Ruihao Gong, Xianglong Liu, Xiangguo Zhang, Haotong Qin, Fengwei Yu, Yuhang Li, Renshuai Tao, Qinghua Yan

Publikováno v: CVPR

Quantization has emerged as one of the most prevalent approaches to compress and accelerate neural networks. Recently, data-free quantization has been widely studied as a practical and promising solution. It synthesizes data for calibrating the quant

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::430fc7b013f5fdb1b31337bdc3354734
https://doi.org/10.1109/cvpr46437.2021.01540

Zobrazit plný text záznamu

Multi-Pretext Attention Network for Few-shot Learning with Self-supervision

Autor: Yifu Ding, Shuo Wang, Renshuai Tao, Xianglong Liu, Hainan Li, Haotong Qin, Jun Li

Few-shot learning is an interesting and challenging study, which enables machines to learn from few samples like humans. Existing studies rarely exploit auxiliary information from large amount of unlabeled data. Self-supervised learning is emerged as

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8108c6a95231e4fba8a4926c9ab722f0
http://arxiv.org/abs/2103.05985

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání