Zhou Daquan

Generative artificial intelligence

I start working on video generation in the year 2022. My research interests are on explainable video representation design and efficient long video generation in terms of both training and inference phases.

Hardware Algorithm Co-design

Hardware Algorithm Co-design is a synergistic approach to developing the hardware and the algorithms in tandem to optimize system performance. I am extremely interested in research projects relating to deep neural network architecture and hardware memory co-design.

Model Efficiency

I keep working on research projects relating to deep neural architecture and dataset efficiency on discrimination, generation and multi-modality models.

Generative Model Design

StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation

Yupeng Zhou, Daquan Zhou*†, Ming-Ming Cheng, Jiashi Feng, Qibin Hou*†

Sora Generates Videos with Stunning Geometrical Consistency

Xuanyi Li, Daquan Zhou, Chenxu Zhang, Shaodong Wei, Qibin Hou, Ming-Ming Cheng

Magic-Me: Identity-Specific Video Customized Diffusion

Ze Ma*, Daquan Zhou*†, Chun-Hsiao Yeh, Xue-She Wang, Xiuyu Li, Huanrui Yang, Zhen Dong †, Kurt Keutzer, Jiashi Feng

MagicVideo: Efficient Video Generation With Latent Diffusion Models

Daquan Zhou, Weimin Wang, Hanshu Yan, Weiwei Lv, Yizhe Zhu, Jiashi Feng.

MagicVideo-V2: Multi-Stage High-Aesthetic Video Generation

Weimin Wang, Jiawei Liu, Zhijie Lin, Jiangqiao Yan, Shuo Chen, Chetwin Low, Tuyen Hoang, Jie Wu, Jun Hao Liew, Hanshu Yan, Daquan Zhou, Jiashi Feng

Difffit: Unlocking transferability of large diffusion models via simple parameter-efficient fine-tuning

Enze Xie, Lewei Yao, Han Shi, Zhili Liu, Daquan Zhou, Zhaoqiang Liu, Jiawei Li, Zhenguo Li. Accepted by Computer Vision and Pattern Recognition Conference (CVPR) 2023

Expanding small-scale datasets with guided imagination

Yifan Zhang, Daquan Zhou*, Bryan Hooi, Kai Wang, Jiashi Feng. Correspondence author, project lead. Accepted by Conference on Neural Information Processing Systems (NeurIPS) 2023

Diffusion probabilistic model made slim

Xingyi Yang, Daquan Zhou, Jiashi Feng, Xinchao Wang. Accepted by Computer Vision and Pattern Recognition Conference (CVPR) 2022

Efficiency

PLLaVA: Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning

Lin Xu, Yilin Zhao, Daquan Zhou*, Zhijie Lin, See Kiong Ng, Jiashi Feng

Dataset Quantization

Daquan Zhou, Kai Wang, Jianyang Gu, Xiangyu Peng, Dongze Lian, Yifan Zhang, Yang You, Jiashi Feng Accepted by ICCV, or the International Conference on Computer Vision (ICCV) 2023. A new dataset compression pipeline with 5× to 10× training speed up.

Scaling & Shiffing Your Features: A New Baseline for Efficient Model Tuning

Dongze Lian*, Daquan Zhou*, Jiashi Feng, Xinchao Wang Accepted by Conference on Neural Information Processing Systems (NeurIPS) 2022, spotlight presentation, Euqual First Author

EPIM: Efficient Processing-In-Memory Accelerators based on Epitome

Chenyu Wang*, Zhen Dong*, Daquan Zhou*, Zhenhua Zhu, Yu Wang, Jiashi Feng, Kurt Keutzer Accepted by Design Automation Conference (DAC) 2024. Equal first author. Collaboration with scholars from UC Berkeley for hardware-algorithm co-design project

Rethinking Bottleneck Structure for Efficient Mobile Network Design

Daquan Zhou, Qibin Hou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan Accepted by the European Conference on Computer Vision (ECCV) 2020, spotlight presentation

Neural Epitome Search for DNN Model Compression

Daquan Zhou, Xiaojie Jin, Qibin Hou, Kaixin Wang, Jianchao Yang, Jiashi Feng Accepted by International Conference on Learning Representations (ICLR) 2020

Coordinate Attention for Efficient Mobile Network Design

• Qibin Hou, Daquan Zhou, Jiashi Feng • Accepted by Computer Vision and Pattern Recognition Conference (CVPR) 2020 • Ranked 5 among Most Influential Paper in CVPR

Deep Model Reassembly

• Xingyi Yang, Daquan Zhou, Songhua Liu, Jingwen Ye, Xinchao Wang • Accepted by Conference on Neural Information Processing Systems (NeurIPS) 2022, best paper nomination.

Convbert: Improving bert with span-based dynamic convolution

• Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan • Accepted by Conference on Neural Information Processing Systems (NeurIPS) 2020

Model Generaliation

DeepViT: Towards Deeper Vision Transformer

• Daquan Zhou, Bingyi Kang, Xiaojie Jin, Linjie Yang, Xiaochen Lian, Qibin Hou, Jiashi Feng • Accepted as a journal paper in IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI) 2022

Understanding The Robustness in Vision Transformers

• Daquan Zhou, Zhiding Yu, Enze Xie, Chaowei Xiao, Anima Anandkumar, Jiashi Feng, Jose M. Alvarez • Accepted by International Conference on Machine Learning (ICML) 2022 (short oral presentation) Champion solutions for Robust Vision Challenge 2022

AutoSpace: Neural Architecture Search with Less Human Interference

• Daquan Zhou, Xiaojie Jin, Xiaochen Lian, Linjie Yang, Yujing Xue, Qibin Hou, Jiashi Feng • Accepted by ICCV, or the International Conference on Computer Vision (ICCV) 2021

Shunted Self-Attention via Multi-Scale Token Aggregation

• Sucheng Ren∗, Daquan Zhou∗, Shengfeng He, Jiashi Feng, Xinchao Wang • Accepted by IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR) 2022, Oral Presentation, Equal First Author

Velocity-to-velocity human motion forecasting

• Hongsong Wang, Liang Wang, Jiashi Feng, Daquan Zhou • Accepted by 2022 IEEE Transactions on Image Processing (TIP) 2022

PANet: Few-Shot Image Semantic Segmentation with Prototype Alignment

• Kaixin Wang, Jun Hao Liew, Yingtian Zou, Daquan Zhou, Jiashi Feng • Accepted by IEEE/CVF International Conference on Computer Vision (ICCV) 2019