Deep Regionlets: Blended Representation and Deep Learning for Generic Object Detection

TL;DR

Deep Regionlets introduces an end-to-end deep learning framework that integrates regionlet-based modeling for flexible, accurate generic object detection, effectively handling deformations and appearance variations.

Contribution

The paper presents a novel deep learning-based object detection method that incorporates non-rectangular region selection and a gating network for adaptive feature pooling, trained end-to-end.

Findings

Achieves competitive results on PASCAL VOC and MS COCO datasets.

Outperforms some state-of-the-art methods like RetinaNet and Mask R-CNN.

Effectively models object deformations and appearance variations.

Abstract

In this paper, we propose a novel object detection algorithm named "Deep Regionlets" by integrating deep neural networks and a conventional detection schema for accurate generic object detection. Motivated by the effectiveness of regionlets for modeling object deformations and multiple aspect ratios, we incorporate regionlets into an end-to-end trainable deep learning framework. The deep regionlets framework consists of a region selection network and a deep regionlet learning module. Specifically, given a detection bounding box proposal, the region selection network provides guidance on where to select sub-regions from which features can be learned from. An object proposal typically contains 3-16 sub-regions. The regionlet learning module focuses on local feature selection and transformations to alleviate the effects of appearance variations. To this end, we first realize non-rectangular region selection within the detection framework to accommodate variations in object appearance. Moreover, we design a "gating network" within the regionlet leaning module to enable instance dependent soft feature selection and pooling. The Deep Regionlets framework is trained end-to-end without additional efforts. We present ablation studies and extensive experiments on the PASCAL VOC dataset and the Microsoft COCO dataset. The proposed method yields competitive performance over state-of-the-art algorithms, such as RetinaNet and Mask R-CNN, even without additional segmentation labels.