A Generic Visualization Approach for Convolutional Neural Networks
This work addresses the need for better visualization tools in retrieval networks, which are essential for searching and indexing, offering a generic method that is incremental but improves upon existing baselines.
The paper tackles the problem of attention visualization for retrieval networks, which is less studied compared to classification networks, by formulating it as a constrained optimization problem using an L2-Norm constraint (L2-CAF) that localizes attention without architectural changes or fine-tuning. It achieves state-of-the-art results on classification networks and significant improvements over a Grad-CAM baseline for retrieval networks.
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