I am a final-year PhD Student at Visual Informatics Group at Texas A&M University, working with Prof. Zhangyang Wang. I received my bachelor's degree from Tongji University.

My research interests lie in the intersection of computer vision and machine learning. In particular, I am interested in enabling efficient machine learning in a broad spectrum of computer vision problems, which includes:

(i) Multimodal Understanding (Modality-agnostic Multimodal Pre-training, Multimodal Fusion)
(ii) Neural Network Design (Neural Architecture Search, Neural Network Compression)
(iii) Computational Photography (Image/ Video Deblurring, Denoising)

Email: sandboxmaster [at] tamu (dot) edu
Google Scholar / GitHub / Linkedin / Resume

I'll be joining Google DeepMind NYC starting Summer 2023.

Experience


Publications (show selected / how by date / show by topic)
Research Topics: Multi-Modal Understanding / Efficient Neural Networks / Image Restoration / Visual Saliency / Biometrics

(* indicates equal contribution)

Junru Wu, Yi Liang, Feng Han, Hassan Akbari, Zhangyang Wang, and Cong Yu
Scaling Multimodal Pre-Training via Cross-Modality Gradient Harmonization
NeurIPS 2022, [Project] [Paper]

Qiucheng Wu*, Yifan Jiang*, Junru Wu*, Kai Wang, Gong Zhang, Humphrey Shi, Zhangyang Wang, Shiyu Chang, and Zhangyang Wang
Grasping the Arrow of Time from the Singularity: Decoding Micromotion in Low-dimensional Latent Spaces from StyleGAN
Arxiv Preprint, [Paper] [Project] [Code] [Demo]

Yifan Jiang, Xinyu Gong, Junru Wu, Humphrey Shi, Zhicheng Yan, and Zhangyang Wang
Auto-X3D: Ultra-Efficient Video Understanding via Finer-Grained Neural Architecture Search
WACV 2022, [Paper]


Junru Wu, Xiang Yu, Ding Liu, Manmohan Chandraker, and Zhangyang Wang
DAVID: Dual-Attentional Video Deblurring
WACV 2020, [Paper]

Rosaura G. VidalMata,... Junru Wu,... Zhangyang Wang,... Walter J. Scheirer
Bridging the gap between computational photography and visual recognition
TPAMI 2020, [Paper] [Code]

Orest Kupyn, Tetiana Martyniuk, Junru Wu and Zhangyang Wang,
DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better
ICCV 2019, [Paper] [Code]
Third-party Adaptation: [OpenVINO] [GIMP-ML] [GIMP-ML(Demo)]

Junru Wu, Xiang Yu, Buyu Liu, Manmohan Chandraker, and Zhangyang Wang
Uncertainty-Aware Physically-Guided Proxy Tasks for Unseen Domain Face Anti-spoofing
Arxiv Preprint, [Paper]

Sicheng Wang, Bihan Wen, Junru Wu, Dacheng Tao, and Zhangyang Wang
Segmentation-Aware Image Denoising without Knowing True Segmentation
Arxiv Preprint, [Paper] [Code]

Junru Wu, Yue Wang, Zhenyu Wu, Zhangyang Wang, Ashok Veeraraghavan, and Yingyan Lin
Deep k-Means: Re-Training and Parameter Sharing with Harder Cluster Assignments for Compressing Deep Convolutions
ICML 2018, [Paper] [Code]

Yanyu Xu, Yanbing Dong, Junru Wu, Zhengzhong Sun, Zhiru Shi, Jingyi Yu, and Shenghua Gao
Gaze Prediction in Dynamic 360◦ Immersive Videos
CVPR 2018, [Paper] [Code]

Yanyu Xu*, Shenghua Gao* Junru Wu*, Nianyi Li, and Jingyi Yu
Personalized Saliency and Its Prediction
TPAMI 2018, [Paper] [Code]

Yanyu Xu, Nianyi Li, Junru Wu, Jingyi Yu, and Shenghua Gao
Beyond Universal Saliency: Personalized Saliency Prediction with Multi-task CNN
IJCAI 2017, [Paper] [Code], Best Student Paper Finalist

Education



Professional Service


* Cartoon portrait credit to Shuai Yang

© Junru Wu 2022