Seeing without Looking: Contextual Rescoring of Object Detections for AP Maximization

TL;DR

This paper introduces a context-based post-processing method using a bidirectional RNN with attention to rescore object detection confidences, improving AP by leveraging detection relationships without visual features.

Contribution

It presents a novel, architecture-agnostic rescoring approach that enhances detection accuracy by modeling contextual information among detections, improving AP without additional visual feature computation.

Findings

Consistent AP improvements over baseline detectors.

Effective reduction of duplicate and out-of-context detections.

Method is computationally inexpensive and architecture-agnostic.

Abstract

The majority of current object detectors lack context: class predictions are made independently from other detections. We propose to incorporate context in object detection by post-processing the output of an arbitrary detector to rescore the confidences of its detections. Rescoring is done by conditioning on contextual information from the entire set of detections: their confidences, predicted classes, and positions. We show that AP can be improved by simply reassigning the detection confidence values such that true positives that survive longer (i.e., those with the correct class and large IoU) are scored higher than false positives or detections with small IoU. In this setting, we use a bidirectional RNN with attention for contextual rescoring and introduce a training target that uses the IoU with ground truth to maximize AP for the given set of detections. The fact that our approach does not require access to visual features makes it computationally inexpensive and agnostic to the detection architecture. In spite of this simplicity, our model consistently improves AP over strong pre-trained baselines (Cascade R-CNN and Faster R-CNN with several backbones), particularly by reducing the confidence of duplicate detections (a learned form of non-maximum suppression) and removing out-of-context objects by conditioning on the confidences, classes, positions, and sizes of the co-occurrent detections. Code is available at https://github.com/LourencoVazPato/seeing-without-looking/