Proposal-Contrastive Pretraining for Object Detection from Fewer Data
This work addresses the challenge of reducing computational costs for unsupervised pretraining in object detection, which is important for researchers and practitioners with limited resources, though it appears incremental as it builds on existing contrastive learning and transformer-based methods.
The paper tackles the problem of resource-intensive unsupervised pretraining for object detection by introducing Proposal Selection Contrast (ProSeCo), which leverages transformer-based detectors to generate object proposals for contrastive learning with smaller batch sizes, achieving state-of-the-art results on standard and novel benchmarks with fewer data.
The use of pretrained deep neural networks represents an attractive way to achieve strong results with few data available. When specialized in dense problems such as object detection, learning local rather than global information in images has proven to be more efficient. However, for unsupervised pretraining, the popular contrastive learning requires a large batch size and, therefore, a lot of resources. To address this problem, we are interested in transformer-based object detectors that have recently gained traction in the community with good performance and with the particularity of generating many diverse object proposals. In this work, we present Proposal Selection Contrast (ProSeCo), a novel unsupervised overall pretraining approach that leverages this property. ProSeCo uses the large number of object proposals generated by the detector for contrastive learning, which allows the use of a smaller batch size, combined with object-level features to learn local information in the images. To improve the effectiveness of the contrastive loss, we introduce the object location information in the selection of positive examples to take into account multiple overlapping object proposals. When reusing pretrained backbone, we advocate for consistency in learning local information between the backbone and the detection head. We show that our method outperforms state of the art in unsupervised pretraining for object detection on standard and novel benchmarks in learning with fewer data.