MOST: Multiple Object localization with Self-supervised Transformers for object discovery
It addresses the problem of unsupervised object localization for computer vision, enabling multiple object detection without labeled data, which is incremental as it builds on existing self-supervised transformer methods.
The paper tackles unsupervised object localization by introducing MOST, which uses self-supervised transformer features and fractal analysis to localize multiple objects per image, outperforming state-of-the-art methods on benchmarks like PASCAL-VOC and COCO20k with improved performance in object detection tasks.
We tackle the challenging task of unsupervised object localization in this work. Recently, transformers trained with self-supervised learning have been shown to exhibit object localization properties without being trained for this task. In this work, we present Multiple Object localization with Self-supervised Transformers (MOST) that uses features of transformers trained using self-supervised learning to localize multiple objects in real world images. MOST analyzes the similarity maps of the features using box counting; a fractal analysis tool to identify tokens lying on foreground patches. The identified tokens are then clustered together, and tokens of each cluster are used to generate bounding boxes on foreground regions. Unlike recent state-of-the-art object localization methods, MOST can localize multiple objects per image and outperforms SOTA algorithms on several object localization and discovery benchmarks on PASCAL-VOC 07, 12 and COCO20k datasets. Additionally, we show that MOST can be used for self-supervised pre-training of object detectors, and yields consistent improvements on fully, semi-supervised object detection and unsupervised region proposal generation.