A Generic Visualization Approach for Convolutional Neural Networks

ECCV 2020

Ahmed Taha      Xitong Yang      Abhinav Shrivastava      Larry Davis

University Of Maryland College Park

Arxiv    TensorFlow    PyTorch

Overview Video


Retrieval networks are essential for searching and indexing. Compared to classification networks, attention visualization for retrieval networks is hardly studied. We formulate attention visualization as a constrained optimization problem. We leverage the unit L2-Norm constraint as an attention filter (L2-CAF) to localize attention in both classification and retrieval networks. Unlike recent literature, our approach requires neither architectural changes nor fine-tuning. Thus, a pre-trained network's performance is never undermined L2-CAF is quantitatively evaluated using weakly supervised object localization. State-of-the-art results are achieved on classification networks. For retrieval networks, significant improvement margins are achieved over a Grad-CAM baseline. Qualitative evaluation demonstrates how the L2-CAF visualizes attention per frame for a recurrent retrieval network. Further ablation studies highlight the computational cost of our approach and compare L2-CAF with other feasible alternatives. Code available at https://bit.ly/3iDBLFv.


title={A Generic Visualization Approach for Convolutional Neural Networks},
author={Taha, Ahmed and Yang, Xitong and Shrivastava, Abhinav and Davis, Larry},
booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},


This work was partially funded by independent grantsfrom Office of Naval Research (N000141612713) and Facebook AI