Research map / Detection and segmentation / Tracking
Online Multi-Target Tracking Using Recurrent Neural Networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Replace Kalman motion estimation with a learned RNN motion model while retaining Hungarian association | Online tracking on MOTChallenge training dataset · MOTChallenge training dataset | MOTA: 19.2 -> 24.0 | Helped |
| Replace Hungarian association with learned LSTM data association (ablation) | RNN tracking on MOTChallenge training dataset · MOTChallenge training dataset | MOTA: 24.0 -> 22.3 | Hurt |
| Use the proposed RNN_LSTM tracker instead of MDP | Online tracking on MOTChallenge test dataset · MOTChallenge test dataset | FPS: 1.1 -> 165.2 | Helped |
| Use the proposed RNN_LSTM tracker instead of MDP | Online tracking on MOTChallenge test dataset · MOTChallenge test dataset | MOTA: 30.3% -> 19.0% | Hurt |
Closest papers by meaning
- Recurrently Target-Attending Tracking (CVPR 2016 · similarity 0.90)
- Learning Multi-Domain Convolutional Neural Networks for Visual Tracking (CVPR 2016 · similarity 0.88)
- Release the Power of Online-Training for Robust Visual Tracking (AAAI 2020 · similarity 0.87)
- Deep Tracking: Seeing Beyond Seeing Using Recurrent Neural Networks (AAAI 2016 · similarity 0.87)
- Target-Aware Deep Tracking (CVPR 2019 · similarity 0.87)
- DASOT: A Unified Framework Integrating Data Association and Single Object Tracking for Online Multi-Object Tracking (AAAI 2020 · similarity 0.87)
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.