Toward Scale-Invariance and Position-Sensitive Region Proposal Networks

TL;DR

This paper introduces a novel object proposal network that enhances scale-invariance and position sensitivity, significantly improving average recall on standard datasets while maintaining real-time performance.

Contribution

The proposed network architecture combines translation-invariance and scale-invariance with large receptive fields, offering a simple yet effective approach for high-quality object proposals.

Findings

Improves AR at 1,000 proposals by 35% on PASCAL VOC

Achieves 45% AR improvement on COCO dataset

Runs at 44.8 ms inference time for 640x640 images

Abstract

Accurately localising object proposals is an important precondition for high detection rate for the state-of-the-art object detection frameworks. The accuracy of an object detection method has been shown highly related to the average recall (AR) of the proposals. In this work, we propose an advanced object proposal network in favour of translation-invariance for objectness classification, translation-variance for bounding box regression, large effective receptive fields for capturing global context and scale-invariance for dealing with a range of object sizes from extremely small to large. The design of the network architecture aims to be simple while being effective and with real time performance. Without bells and whistles the proposed object proposal network significantly improves the AR at 1,000 proposals by $35\%$ and $45\%$ on PASCAL VOC and COCO dataset respectively and has a fast inference time of 44.8 ms for input image size of $640^{2}$. Empirical studies have also shown that the proposed method is class-agnostic to be generalised for general object proposal.