AI Paper Atlas

Research map / Detection and segmentation / Tracking

Detection and Tracking Meet Drones Challenge

Pengfei Zhu, Longyin Wen, Dawei Du, Xiao Bian, Heng Fan, Qinghua Hu, Haibin Ling

TPAMI 2021 · Detection and segmentation › Tracking · approach: Unified transformer tracking · 1,587 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use video object detectors with temporal information instead of image detectorsD&T vs FPN · VisDrone-VID test-devAP: 12.93 -> 14.21Improved
combine several effective modules on the Cascade R-CNN baselineCascade R-CNN · VisDrone-DET test-challenge 2020AP: 16.09 -> 34.57Improved
integrate deformable convolution and global context into Cascade R-CNNCascade R-CNN · VisDrone-VID test-challenge 2019AP: 16.09 -> 29.22Improved
train on large-scale datasets offline for trackersATOM · VisDrone-SOT test-devsuccess score: 64.5Improved
shift from correlation filter to deep network trackersVisDrone-SOT winners · VisDrone-SOT test-challenge 2018success score: 68.0 -> 75.5Improved
ensemble several trackers (ATOM, SiamRPN++, Siam R-CNN, DiMP)LTNMI · VisDrone-SOT test-challenge 2018success score: 75.5 -> 76.5Improved
add squeeze-and-excitation to ATOMATOM · VisDrone-SOT test-challenge 2018success score: 75.5Improved
use newly collected long-term sequences (2019 set)best SOT trackers · VisDrone-SOTsuccess score: 68.0 -> 48.9Hurt
improve trackers from 2019 to 2020 challengebest SOT tracker · VisDrone-SOT test-challenge 2019success score: 48.9 -> 55.5Improved
use video detector over detector improvement via feature enhancement (SSD)CFE-SSDv2 · VisDrone-VID test-challenge 2018AP: 8.63 -> 21.57Improved
add tracking (ECO) to associate detectionsSSD · VisDrone-VID test-challenge 2018AP: 21.57 -> 16.54Mixed
remove higher convolutional layers and add attention to RetinaNetRetinaNet · VisDrone-DET test-challenge 2018AP: 18.94 -> 31.88Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy