Mrigank Rochan

Department of Computer Science
176 Thorvaldson Bldg
110 Science Place
University of Saskatchewan
Saskatoon, SK S7N 5C9, Canada

Email / Google Scholar

About Me

I am an Assistant Professor in the Department of Computer Science at the University of Saskatchewan. Previously, I was a Senior Researcher in the Autonomous Driving Perception team at Huawei Noah's Ark Lab in Toronto. I received my Ph.D. and M.Sc. in Computer Science from the University of Manitoba in 2020 and 2016, respectively, where I was advised by Prof. Yang Wang. I obtained my B.Tech. degree in Computer Science and Engineering from Amrita University, India, in 2011. I was a visiting research student at Simon Fraser University in 2015-2016. During my Ph.D., I also did a research internship at Mapillary Research (acquired by Meta).

My research is focused on Deep Learning and its applications in Computer Vision. For my thesis, I received the 2020 CIPPRS John Barron Doctoral Dissertation Award, a prestigious national award given annually by the Canadian Image Processing and Pattern Recognition Society (CIPPRS) to the top Ph.D. thesis in computer or robot vision.

Prospective Students: I am seeking graduate students (MSc and PhD) to join my research group in Fall 2026. Students with experience in computer vision and deep learning are especially encouraged to apply (deadline is December 15, 2025). Please mention my name as a preferred supervisor in your application. You do not need my permission to select my name as a preferred supervisor. If there is a potential fit, I will contact you. For details on the application process and funding, please visit our graduate admissions page.

News

^NEW 09/2025: One paper accepted to WACV 2026 in Round 1 (acceptance rate: 85/1329 = 6.4%)!
^NEW 08/2025: One paper accepted to EMNLP 2025!
^NEW 10/2024: One paper accepted to WACV 2025!
^NEW 10/2024: Two papers accepted to self-supervised learning workshop at NeurIPS 2024!
^NEW 07/2024: Received the Google PaliGemma Academic Program Award!
^NEW 06/2024: Awarded the NSERC Discovery Grant!
03/2024: Two papers accepted to IEEE IV 2024!
03/2023: One paper accepted to IEEE IV 2023!
03/2022: One paper accepted to CVPR 2022!
01/2022: One paper accepted to ICRA 2022!
07/2021: One paper accepted to ICCV 2021!
01/2021: One paper accepted to IEEE TPAMI!

Invited Talks

^NEW 05/2025: Invited symposium speaker at the Conference on Robots and Vision (CRV), 2025!
^NEW 10/2024: Talk at Adobe Research!
^NEW 10/2024: Talk at Microsoft Research!
12/2023: Talk at Google Research!
12/2023: Talk at Indian Institute of Science (IISc), Bangalore!

Teaching

CMPT 489/828: Deep Learning [Winter 2026, Winter 2025, Winter 2024]
CMPT 318: Data Analytics [Winter 2026, Winter 2025]

Selected Publications

	Test-Time Adaptation for Video Highlight Detection Using Meta-Auxiliary Learning and Cross-Modality Hallucinations Zahidul Islam, Sujoy Paul, and Mrigank Rochan IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2026. [arXiv]
	AdaptMerge: Inference Time Adaptive Visual and Language-Guided Token Merging for Efficient Large Multimodal Models Zahidul Islam and Mrigank Rochan Findings of Empirical Methods in Natural Language Processing (EMNLP), 2025. [paper]
	Motion-Focused Tokenization for Source-Free Video Domain Adaptation Tzu Ling Liu, Ian Stavness, and Mrigank Rochan Tokenization Workshop, International Conference on Machine Learning (ICML), 2025. [paper]
	Unsupervised Video Highlight Detection by Learning from Audio and Visual Recurrence Zahidul Islam, Sujoy Paul, and Mrigank Rochan IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2025. Workshop on Self-Supervised Learning, Neural Information Processing Systems (NeurIPS), 2024. [paper]
	Large Language Models as Robust Data Generators in Software Analytics: Are We There Yet? Md Awal, Mrigank Rochan, and Chanchal Roy AI Models/Data track at the International Conference on Evaluation and Assessment in Software Engineering (EASE), 2025. [arXiv]
	Test-Time Adaptation for Video Highlight Detection Zahidul Islam, Sujoy Paul, and Mrigank Rochan Workshop on Self-Supervised Learning, Neural Information Processing Systems (NeurIPS), 2024. [paper]
	Gradual Batch Alternation for Effective Domain Adaptation in LiDAR-Based 3D Object Detection Mrigank Rochan, Xingxin Chen, Alaap Grandhi, Eduardo Corral-Soto, and Bingbing Liu IEEE Intelligent Vehicles Symposium (IV), 2024. [paper]
	Effects of Range-based LiDAR Point Cloud Density Manipulation on 3D Object Detection Eduardo Corral-Soto, Alaap Grandhi, Yannis He, Mrigank Rochan, and Bingbing Liu IEEE Intelligent Vehicles Symposium (IV), 2024. [paper]
	Domain Adaptation in LiDAR Semantic Segmentation via Hybrid Learning With Alternating Skip Connections Eduardo Corral-Soto, Mrigank Rochan, Yannis He, Xingxin Chen, Shubhra Aich, and Bingbing Liu IEEE Intelligent Vehicles Symposium (IV), 2023. [paper]
	Contrastive Learning for Unsupervised Video Highlight Detection Taivanbat Badamdorj, Mrigank Rochan, Yang Wang, and Li Cheng IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022. [paper]
	Unsupervised Domain Adaptation in LiDAR Semantic Segmentation with Self-Supervision and Gated Adapters Mrigank Rochan, Shubhra Aich, Eduardo Corral-Soto, Amir Nabatchian, and Bingbing Liu IEEE International Conference on Robotics and Automation (ICRA), 2022. [paper]
	Joint Visual and Audio Learning for Video Highlight Detection Taivanbat Badamdorj, Mrigank Rochan, Yang Wang, and Li Cheng IEEE/CVF International Conference on Computer Vision (ICCV), 2021. [paper]
	Referring Segmentation in Images and Videos with Cross-Modal Self-Attention Network Linwei Ye, Mrigank Rochan, Zhi Liu, Xiaoqin Zhang, and Yang Wang IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021. [arXiv] [code]
	AdaCrowd: Unlabeled Scene Adaptation for Crowd Counting Mahesh Reddy, Mrigank Rochan, Yiwei Lu, and Yang Wang IEEE Transactions on Multimedia (TMM), 2021. [arXiv] [code]
	Efficient Deep Learning Models for Video Abstraction Mrigank Rochan Ph.D. Thesis, University of Manitoba, 2020. CIPPRS John Barron Doctoral Dissertation Award [announcement] [link]
	Adaptive Video Highlight Detection by Learning from User History Mrigank Rochan, Mahesh Reddy, Linwei Ye, and Yang Wang European Conference on Computer Vision (ECCV), 2020. [arXiv] [code]
	Sentence Guided Temporal Modulation for Dynamic Video Thumbnail Generation Mrigank Rochan, Mahesh Reddy, and Yang Wang British Machine Vision Conference (BMVC), 2020. [arXiv]
	Few-Shot Scene Adaptive Crowd Counting Using Meta-Learning Mahesh Reddy, Md Hossain, Mrigank Rochan and Yang Wang IEEE Winter Conference of Applications on Computer Vision (WACV), 2020. [arXiv] [code]
	Video Summarization by Learning from Unpaired Data Mrigank Rochan and Yang Wang IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. [arXiv]
	Cross-Modal Self-Attention Network for Referring Image Segmentation Linwei Ye, Mrigank Rochan, Zhi Liu, and Yang Wang IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. [arXiv] [code]
	Convolutional Temporal Attention Model for Video-based Person Re-identification Tanzila Rahman, Mrigank Rochan, and Yang Wang IEEE International Conference on Multimedia and Expo (ICME), 2019. [arXiv]
	Video Summarization Using Fully Convolutional Sequence Networks Mrigank Rochan, Linwei Ye, and Yang Wang European Conference on Computer Vision (ECCV), 2018. [paper] [arXiv]
	Future Semantic Segmentation with Convolutional LSTM Shahab Nabavi, Mrigank Rochan, and Yang Wang British Machine Vision Conference (BMVC), 2018. [arXiv]
	Gated Feedback Refinement Network for Dense Image Labeling Md Amirul Islam, Mrigank Rochan, Neil Bruce, and Yang Wang IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. [paper] [extended version] [code]
	Person Re-Identification by Localizing Discriminative Regions Tanzila Rahman, Mrigank Rochan, and Yang Wang British Machine Vision Conference (BMVC), 2017. [paper]
	Adapting Object Detectors from Images to Weakly Labeled Videos Omit Chanda, Eu Wern Teh, Mrigank Rochan, Zhenyu Guo, and Yang Wang British Machine Vision Conference (BMVC), 2017. [paper] [code]
	Salient Object Detection using a Context-Aware Refinement Network Md Amirul Islam, Mahmoud Kalash, Mrigank Rochan, Neil Bruce, and Yang Wang British Machine Vision Conference (BMVC), 2017. [paper]
	Label Refinement Network for Coarse-to-Fine Semantic Segmentation Md Amirul Islam, Shujon Naha, Mrigank Rochan, Neil Bruce, and Yang Wang arXiv preprint arXiv:1606.07415, 2017. [arXiv]
	Attention Networks for Weakly Supervised Object Localization Eu Wern Teh, Mrigank Rochan, and Yang Wang British Machine Vision Conference (BMVC), 2016. [paper]
	Object Localization in Weakly Labeled Images and Videos Mrigank Rochan M.Sc. Thesis, University of Manitoba, 2016. [link]
	Weakly Supervised Object Localization and Segmentation in Videos Mrigank Rochan and Yang Wang Image and Vision Computing (IVC), 2016. [paper]
	Weakly Supervised Localization of Novel Objects using Appearance Transfer Mrigank Rochan and Yang Wang IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015. [paper]
	Latent SVM for Object Localization in Weakly Labeled Videos Mrigank Rochan and Yang Wang Canadian Conference on Computer and Robot Vision (CRV), 2015. [paper]
	Efficient Object Localization and Segmentation in Weakly Labeled Videos Mrigank Rochan and Yang Wang International Symposium on Visual Computing (ISVC), 2014. Oral Presentation [paper]
	Segmenting Objects in Weakly Labeled Videos Mrigank Rochan, Shafin Rahman, Neil Bruce, and Yang Wang Canadian Conference on Computer and Robot Vision (CRV), 2014. Oral Presentation [paper] [demo video]
	Examining Visual Saliency Prediction in Naturalistic Scenes Shafin Rahman, Mrigank Rochan, Yang Wang, and Neil Bruce IEEE International Conference on Image Processing (ICIP), 2014. Oral Presentation [paper]
	See Google Scholar for full publication list