Multi-scale Volumes for Deep Object Detection and Localization
This addresses the problem of detecting objects with large scale variation and small objects in computer vision, but it is incremental as it builds on existing multi-scale reasoning approaches.
The study tackled object detection and localization by proposing a multi-scale framework operating on scale volumes of a deep feature pyramid, resulting in significant gains in detection performance and localization quality on datasets like PASCAL VOC and a highway vehicles dataset.
This study aims to analyze the benefits of improved multi-scale reasoning for object detection and localization with deep convolutional neural networks. To that end, an efficient and general object detection framework which operates on scale volumes of a deep feature pyramid is proposed. In contrast to the proposed approach, most current state-of-the-art object detectors operate on a single-scale in training, while testing involves independent evaluation across scales. One benefit of the proposed approach is in better capturing of multi-scale contextual information, resulting in significant gains in both detection performance and localization quality of objects on the PASCAL VOC dataset and a multi-view highway vehicles dataset. The joint detection and localization scale-specific models are shown to especially benefit detection of challenging object categories which exhibit large scale variation as well as detection of small objects.