# CaTDet: Cascaded Tracked Detector for Efficient Object Detection from Video

Source: [CaTDet: Cascaded Tracked Detector for Efficient Object Detection from Video](<https://www.sota2.com/research/paper/68fa656a1e209131cea0a369>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:26:52.925Z

Authors: Huizi Mao, Taeyoung Kong, William J. Dally

Published: 2018-09-30T17:59:42.000Z

## Abstract

Detecting objects in a video is a compute-intensive task. In this paper we propose CaTDet, a system to speedup object detection by leveraging the temporal correlation in video. CaTDet consists of two DNN models that form a cascaded detector, and an additional tracker to predict regions of interests based on historic detections. We also propose a new metric, mean Delay(mD), which is designed for latency-critical video applications. Experiments on the KITTI dataset show that CaTDet reduces operation count by 5.1-8.7x with the same mean Average Precision(mAP) as the single-model Faster R-CNN detector and incurs additional delay of 0.3 frame. On CityPersons dataset, CaTDet achieves 13.0x reduction in operations with 0.8% mAP loss.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1810.00434>)

[PDF](<https://arxiv.org/pdf/1810.00434.pdf>)