I am currently a post-doctoral researcher in CyLabECE Department, CMU, advised by Prof. Marios Savvides. My research interests span computer vision, efficient deep learning, etc.

Prior to CMU, I was fortunate to be a joint-training Ph.D student (2017-2019) in UIUC/IFP group, advised by Prof. Thomas S. Huang.

I received my Ph.D from Fudan University working with Prof. Yu-Gang Jiang and Prof. Xiangyang Xue. I was a research intern at Intel Labs China and SenseTime Research, collaborating with Zhuang Liu, Jianguo Li, Yurong Chen (ILC, 2016-2017), Jianping Shi (SenseTime Research).

Department of Electrical and Computer Engineering, Carnegie Mellon University
5000 Forbes Avenue, Pittsburgh, PA, 15213
Email: zhiqians AT andrew.cmu.edu | zhiqiangshen0214 AT gmail.com
shen54 AT illinois.edu | zhiqiangshen13 AT fudan.edu.cn
[Google Scholar] |  [Github]

Research Interest

My research focuses on the broad areas of machine learning, deep learning and their applications on computer vision. Specifically, I am interested in deep learning methods for object detection, fine-grained recognition, image/video captioning, domain adaption, etc. Recently, I focus on

  • Low-bit Networks
  • Knowledge Distillation
  • Designing and Training High-efficiency Network Structures
  • Weakly-supervised/Un(Self-)supervised Learning
  • Image Understanding, Including Object Detection, Captioning and Fine-grained Recognition
  • Few-shot and Zero-shot Learning


Publications (Full List)

Zhiqiang Shen, Zechun Liu, Zhuang Liu, Marios Savvides, Trevor Darrell, Eric Xing.
Un-Mix: Rethinking Image Mixtures for Unsupervised Visual Representation Learning
Technical Report.
Code & Models  |  arXiv Paper

Zhiqiang Shen, Marios Savvides.
MEAL V2: Boosting Vanilla ResNet-50 to 80%+ Top-1 Accuracy on ImageNet without Tricks
Technical Report. Short version has been accepted in NeurIPS 2020 Beyond BackPropagation: Novel Ideas for Training Neural Architectures workshop.
Code & Models  |  arXiv Paper
We achieve 80.67% top-1 accuracy using a single crop-size of 224×224 on the vanilla ResNet-50, the first work that is able to boost vanilla ResNet-50 to surpass 80% on ImageNet without architecture modification or additional training data. Our result can be regarded as a new strong baseline on ResNet-50 using knowledge distillation.

Zhiqiang Shen, Zechun Liu, Jie Qin, Lei Huang, Kwang-Ting Cheng, Marios Savvides.
S2-BNN: Bridging the Gap Between Self-Supervised Real and 1-bit Neural Networks via Guided Distribution Calibration
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
Self-supervised BNNs using distillation loss (5.5~15% improvement over contrastive baseline).
Code & Models (coming soon)  |  arXiv Paper

Yutong Zheng, Yu-Kai Huang, Ran Tao, Zhiqiang Shen, Marios Savvides.
Unsupervised Disentanglement of Linear-Encoded Facial Semantics
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
arXiv Paper

Chenchen Zhu, Fangyi Chen, Uzair Ahmed, Zhiqiang Shen, Marios Savvides.
Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
arXiv Paper

Zhiqiang Shen, Zechun Liu, Dejia Xu, Zitian Chen, Kwang-Ting Cheng, Marios Savvides.
Is Label Smoothing Truly Incompatible with Knowledge Distillation: An Empirical Study
International Conference on Learning Representations (ICLR), 2021.
OpenReview (Rating: 8 6 6 6)  |  Project Page  |  Paper
A new perspective on the relationship between knowledge distillation and label smoothing.

Zhiqiang Shen*, Zechun Liu*, Jie Qin, Marios Savvides, Kwang-Ting Cheng.
Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning
Association for the Advancement of Artificial Intelligence (AAAI), 2021.
A searching based method for few-shot learning.

Zhiqiang Shen, Mingyang Huang, Jianping Shi, Zechun Liu, Harsh Maheshwari, Yutong Zheng, Xiangyang Xue, Marios Savvides, Thomas S. Huang.
CDTD: A Large-Scale Cross-Domain Benchmark for Instance-Level Image-to-Image Translation and Domain Adaptive Object Detection
International Journal of Computer Vision (IJCV), 2020.
Code & Models  |  Paper

Zechun Liu, Zhiqiang Shen(✝), Marios Savvides, Kwang-Ting Cheng. (✝: Corresponding Author)
ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions
European Conference on Computer Vision (ECCV), 2020.
Code & Models  |  Paper
We achieve 65.9% (ResNet-based) and 69.5% (MobileNet-based) top-1 accuracy on ImageNet (the new results are slightly higher than those in our original paper after fixing a small loading bug), for the first time, exceeding benchmarking ResNet-level accuracy (69.3%), meanwhile, achieving more than 22× reduction in computational complexity.

Chenchen Zhu, Fangyi Chen, Zhiqiang Shen, Marios Savvides.
Soft anchor-point object detection
European Conference on Computer Vision (ECCV), 2020.

Devesh Walawalkar, Zhiqiang Shen, Marios Savvides.
Online Ensemble Model Compression using Knowledge Distillation
European Conference on Computer Vision (ECCV), 2020.

Hai Phan*, Zechun Liu*, Dang Huynh, Marios Savvides, Kwang-Ting Cheng, Zhiqiang Shen.
Binarizing MobileNet via Evolution-based Searching
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

Hai Phan, Dang Huynh, Yihui He, Marios Savvides, Zhiqiang Shen.
MoBiNet: A Mobile Binary Network for Image Classification
IEEE Winter Conference on Applications of Computer Vision (WACV), 2020. (Oral)

Zhuang Liu, Tinghui Zhou, Zhiqiang Shen, Bingyi Kang, Trevor Darrell.
Transferable Recognition-Aware Image Processing
arXiv preprint arXiv:1910.09185 (2019).
arXiv Paper

Zhiqiang Shen, Zhankui He, Wanyun Cui, Jiahui Yu, Yutong Zheng, Chenchen Zhu, Marios Savvides.
Adversarial-Based Knowledge Distillation for Multi-Model Ensemble and Noisy Data Refinement
Extended version of our MEAL. arXiv preprint arXiv:1908.08520 (2019).
arXiv Paper

Zhiqiang Shen, Zhuang Liu, Jianguo Li, Yu-Gang Jiang, Yurong Chen, Xiangyang Xue.
Object Detection from Scratch with Deep Supervision
IEEE transactions on pattern analysis and machine intelligence (T-PAMI), 2019.
Code & Models  |  arXiv Paper

Zhiqiang Shen, Mingyang Huang, Jianping Shi, Xiangyang Xue, Thomas S. Huang.
Towards Instance-level Image-to-Image Translation
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
Project  |  Paper  |  Dataset

Zhiqiang Shen*, Zhankui He*, Xiangyang Xue.
MEAL: Multi-Model Ensemble via Adversarial Learning
33rd AAAI Conference on Artificial Intelligence (AAAI), 2019. (Oral)
Code & Models  |  Our ResNet-50 (Top-1/5: 21.70%/5.99%)   [PyTorch Model (102.5M)]

Wanyun Cui, Guangyu Zheng, Zhiqiang Shen, Sihang Jiang, Wei Wang.
Transfer Learning for Sequences via Learning to Collocate
International Conference on Learning Representations (ICLR), 2019.
OpenReview  |  arXiv Paper

Zhiqiang Shen, Honghui Shi, Jiahui Yu, Hai Phan, Rogerio Feris, Liangliang Cao, Ding Liu, Xinchao Wang, Thomas Huang, Marios Savvides.
Improving Object Detection from Scratch via Gated Feature Reuse
30th British Machine Vision Conference (BMVC), 2019.

Yunchao Wei, Zhiqiang Shen*, Bowen Cheng*, Honghui Shi, Jinjun Xiong, Jiashi Feng, Thomas Huang.
TS2C: Tight Box Mining with Surrounding Segmentation Context for Weakly Supervised Object Detection
European Conference on Computer Vision (ECCV), 2018. (* indicates equal contribution)
arXiv Paper

Zhuang Liu, Jianguo Li, Zhiqiang Shen, Gao Huang, Shoumeng Yan, Changshui Zhang.
Learning Efficient Convolutional Networks through Network Slimming
Proceedings of 16th IEEE International Conference on Computer Vision (ICCV2017).
Code & Models  |  Paper

Zhiqiang Shen*, Zhuang Liu*, Jianguo Li, Yu-Gang Jiang, Yurong Chen, Xiangyang Xue.
DSOD: Learning Deeply Supervised Object Detectors from Scratch
Proceedings of 16th IEEE International Conference on Computer Vision (ICCV2017).
(* indicates equal contribution)
Code & Models  |  Paper

Zhiqiang Shen, Jianguo Li, Zhou Su, Minjun Li, Yurong Chen, Yu-Gang Jiang, Xiangyang Xue.
Weakly Supervised Dense Video Captioning
Proceedings of 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR2017).
Project  |  Paper

Zhiqiang Shen, Yu-Gang Jiang, Dequan Wang, Xiangyang Xue.
Iterative Object and Part Transfer for Fine-Grained Recognition
Proceedings of 18th IEEE Conference on Multimedia & Expo (ICME2017). (Oral)

Dequan Wang, Zhiqiang Shen, Jie Shao, Wei Zhang, Xiangyang Xue, Zheng Zhang.
Multiple granularity descriptors for fine-grained categorization
Proceedings of 15th IEEE International Conference on Computer Vision (ICCV2015).

Dequan Wang, Tianjun Xiao, Zhiqiang Shen, Xiangyang Xue.
Learning Regions and Descriptors for Fine-grained Recognition
Proceedings of 28th IEEE Conference on Computer Vision and Pattern Recognition (CVPR2015 FGVC3 Workshop).

Academic Activities

  • Conference reviewer: ICML 2021, CVPR 2021, AAAI 2021, WACV 2021, NeurIPS 2020, ECCV 2020, BMVC 2020, IJCAI 2020, CVPR 2020, AAAI 2020, ICCV 2019, CVPR 2019, AAAI 2019, CVPR 2018, ACCV 2018, NIPS 2016.
  • Journal reviewer: TPAMI, IJCV, TIP, TMM, JVCI.

Awards and Honors

  • CVPR 2019 Doctoral Consortium travel award. Mentor: Prof. Trevor Darrell.
  • ICLR 2019 travel award, 2019
  • AAAI 2019 student scholarship award, 2018
  • ICCV 2017 student volunteer, 2017
  • Huawei scholarship, 2017
  • During my internship, our team won the 2016 Intel China Award (ICA), the highest award for team achievement in Intel China, 2016
  • Tung OOCL scholarship, 2015
  • Special Grade Scholarship, 2013
  • University-level Outstanding Students, 2013


  • MSR-VTT Challenge (video captioning): ranked 4th in human evaluation and ranked 5th in the automatic evaluation metrics (Team leader), 2016
  • Top 10% in Kaggle Competition of Right Whale Recognition, 2016
  • Second Prize in DataCastle Competition of the Verification Code Recognition, 2016
  • Second Prize (National-level) in China Graduate Student Mathematical Contest in Modeling, 2015
  • MCM/ICM -- Honorable Mention, 2012
  • First Prize (National-level) in Electrical Engineering Mathematical Contest in Modeling, 2012
  • First Prize (National-level) in China Undergraduate Mathematical Contest in Modeling, 2011

Teaching Assistant

  • 2015.9- 2016.1, Fudan University, COMP120008.02, C++ language programming