IMMVP: An Efficient Daytime and Nighttime On-Road Object Detector

TL;DR

This paper introduces IMMVP, a lightweight, multi-condition object detection method that improves accuracy in daytime and nighttime scenarios using subclassing, sample skipping, external data, and a ResNet-18 FPN backbone on edge devices.

Contribution

It presents a novel combination of techniques including subclassing, sample skipping, external data, and a lightweight backbone for robust on-road object detection in varying lighting conditions.

Findings

Improved detection accuracy across day and night conditions.

Effective lightweight model suitable for edge devices.

Enhanced bounding box refinement with Cascade R-CNN.

Abstract

It is hard to detect on-road objects under various lighting conditions. To improve the quality of the classifier, three techniques are used. We define subclasses to separate daytime and nighttime samples. Then we skip similar samples in the training set to prevent overfitting. With the help of the outside training samples, the detection accuracy is also improved. To detect objects in an edge device, Nvidia Jetson TX2 platform, we exert the lightweight model ResNet-18 FPN as the backbone feature extractor. The FPN (Feature Pyramid Network) generates good features for detecting objects over various scales. With Cascade R-CNN technique, the bounding boxes are iteratively refined for better results.