Current browse context:
cs.CV
Change to browse by:
References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: Efficient Human-in-the-loop System for Guiding DNNs Attention
(Submitted on 13 Jun 2022 (v1), last revised 6 Feb 2023 (this version, v3))
Abstract: Attention guidance is an approach to addressing dataset bias in deep learning, where the model relies on incorrect features to make decisions. Focusing on image classification tasks, we propose an efficient human-in-the-loop system to interactively direct the attention of classifiers to the regions specified by users, thereby reducing the influence of co-occurrence bias and improving the transferability and interpretability of a DNN. Previous approaches for attention guidance require the preparation of pixel-level annotations and are not designed as interactive systems. We present a new interactive method to allow users to annotate images with simple clicks, and study a novel active learning strategy to significantly reduce the number of annotations. We conducted both a numerical evaluation and a user study to evaluate the proposed system on multiple datasets. Compared to the existing non-active-learning approach which usually relies on huge amounts of polygon-based segmentation masks to fine-tune or train the DNNs, our system can save lots of labor and money and obtain a fine-tuned network that works better even when the dataset is biased. The experiment results indicate that the proposed system is efficient, reasonable, and reliable.
Submission history
From: Haoran Xie [view email][v1] Mon, 13 Jun 2022 09:04:32 GMT (18232kb,D)
[v2] Tue, 14 Jun 2022 07:53:49 GMT (18230kb,D)
[v3] Mon, 6 Feb 2023 13:11:44 GMT (21799kb,D)
Link back to: arXiv, form interface, contact.