# A Delay Metric for Video Object Detection: What Average Precision Fails to Tell

arXiv: https://arxiv.org/abs/1908.06368 · HTML: https://mlsanity.com/paper/1908.06368/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Aug 2019.

## Built on by (most cited first, 9 of 9)

- [A Modular and Unified Framework for Detecting and Localizing Video Anomalies](https://mlsanity.com/paper/2103.11299/) (2021)
- [Motion Vector Extrapolation for Video Object Detection](https://mlsanity.com/paper/2104.08918/) (2021)
- [An Efficient Approach for Anomaly Detection in Traffic Videos](https://mlsanity.com/paper/2104.09758/) (2021)
- [PLM-Net: Perception Latency Mitigation Network for Vision-Based Lateral Control of Autonomous Vehicles](https://mlsanity.com/paper/2407.16740/) (2024)
- [Single-shot Star-convex Polygon-based Instance Segmentation for Spatially-correlated Biomedical Objects](https://mlsanity.com/paper/2504.12078/) (2025)
- [Identification of Driver Phone Usage Violations via State-of-the-Art Object Detection with Tracking](https://mlsanity.com/paper/2109.02119/) (2021)
- [Checklist to Define the Identification of TP, FP, and FN Object Detections in Automated Driving](https://mlsanity.com/paper/2308.07106/) (2023)
- [Perception Characteristics Distance: Measuring Stability and Robustness of Perception System in Dynamic Conditions under a Certain Decision Rule](https://mlsanity.com/paper/2506.09217/) (2025)
- [Neuromorphic Object Detection: An In-Depth Study and Future Directions](https://mlsanity.com/paper/2607.23576/) (2026)

## Abstract (arXiv)

Average precision (AP) is a widely used metric to evaluate detection accuracy of image and video object detectors. In this paper, we analyze object detection from videos and point out that AP alone is not sufficient to capture the temporal nature of video object detection. To tackle this problem, we propose a comprehensive metric, average delay (AD), to measure and compare detection delay. To facilitate delay evaluation, we carefully select a subset of ImageNet VID, which we name as ImageNet VIDT with an emphasis on complex trajectories. By extensively evaluating a wide range of detectors on VIDT, we show that most methods drastically increase the detection delay but still preserve AP well. In other words, AP is not sensitive enough to reflect the temporal characteristics of a video object detector. Our results suggest that video object detection methods should be additionally evaluated with a delay metric, particularly for latency-critical applications such as autonomous vehicle perception.