TALISMAN: Targeted Active Learning for Object Detection with Rare Classes and Slices using Submodular Mutual Information

TL;DR

TALISMAN is a targeted active learning framework that efficiently identifies and acquires data with rare slices for object detection, improving performance on critical but infrequent data subsets.

Contribution

It introduces a novel submodular mutual information-based approach for targeted active learning focused on rare data slices in object detection.

Findings

Outperforms existing methods on rare slices in PASCAL VOC and BDD100K datasets.

Improves average precision and mAP on critical rare data slices.

Effective in real-world self-driving datasets.

Abstract

Deep neural networks based object detectors have shown great success in a variety of domains like autonomous vehicles, biomedical imaging, etc. It is known that their success depends on a large amount of data from the domain of interest. While deep models often perform well in terms of overall accuracy, they often struggle in performance on rare yet critical data slices. For example, data slices like "motorcycle at night" or "bicycle at night" are often rare but very critical slices for self-driving applications and false negatives on such rare slices could result in ill-fated failures and accidents. Active learning (AL) is a well-known paradigm to incrementally and adaptively build training datasets with a human in the loop. However, current AL based acquisition functions are not well-equipped to tackle real-world datasets with rare slices, since they are based on uncertainty scores or global descriptors of the image. We propose TALISMAN, a novel framework for Targeted Active Learning or object detectIon with rare slices using Submodular MutuAl iNformation. Our method uses the submodular mutual information functions instantiated using features of the region of interest (RoI) to efficiently target and acquire data points with rare slices. We evaluate our framework on the standard PASCAL VOC07+12 and BDD100K, a real-world self-driving dataset. We observe that TALISMAN outperforms other methods by in terms of average precision on rare slices, and in terms of mAP.