Research map / Detection and segmentation / Open-set and few-shot detection
Self-Supervised Moving Vehicle Tracking With Stereo Sound
Chuang Gan, Hang Zhao, Peihao Chen, David Cox, Antonio Torralba
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add camera meta-data input (ablation) | StereoSoundNet · Auditory Vehicle Tracking test | AP@0.5: 27.76 -> 57.47 | Improved |
| add feature alignment loss (ablation) | StereoSoundNet · Auditory Vehicle Tracking test | AP@Ave: 21.55 -> 21.55 | No clear effect |
| add feature alignment loss (ablation) | StereoSoundNet · Auditory Vehicle Tracking test | CD_x: 10.82% -> 10.53% | Improved |
| use stereo sound instead of mono sound (ablation) | StereoSoundNet · Auditory Vehicle Tracking test | AP@0.5: 38.57 -> 57.47 | Improved |
| use spectrogram instead of raw waveform input (ablation) | StereoSoundNet · Auditory Vehicle Tracking test | AP@0.5: 23.14 -> 57.47 | Improved |
| add tracking post-processing | StereoSoundNet · Auditory Vehicle Tracking test | AP@Ave: 21.55 -> 25.05 | Improved |
| add tracking post-processing | StereoSoundNet · Auditory Vehicle Tracking test | MOTA: 13.9% -> 18.7% | Improved |
| add tracking post-processing | StereoSoundNet · Auditory Vehicle Tracking test | ID switches: 1100 -> 954 | Improved |
| add tracking post-processing | StereoSoundNet · Auditory Vehicle Tracking test | Fragments: 1318 -> 502 | Improved |
| use audio student instead of vision YOLOv2 under night lighting | StereoSoundNet · Night videos (5 videos) | AP@0.5: 6.78 -> 30.88 | Improved |
| use camera meta-data for novel scenes (ablation) | StereoSoundNet · Novel scenes (10 train / 5 test scenes) | AP@0.5: 0.00 -> 42.79 | Improved |
| go from single vehicle to multiple vehicles | StereoSoundNet · Auditory Vehicle Tracking test | AP@0.5: 70.12 -> 32.06 | Hurt |
Closest papers by meaning
- Self-Supervised Object Detection From Audio-Visual Correspondence (CVPR 2022 · similarity 0.89)
- Sound Localization by Self-Supervised Time Delay Estimation (ECCV 2022 · similarity 0.88)
- There Is More Than Meets the Eye: Self-Supervised Multi-Object Detection and Tracking With Sound by Distilling Multimodal Knowledge (CVPR 2021 · similarity 0.88)
- Self-Supervised Learning of Audio-Visual Objects from Video (ECCV 2020 · similarity 0.87)
- Supervising Sound Localization by In-the-wild Egomotion (CVPR 2025 · similarity 0.87)
- Prompting Segmentation with Sound Is Generalizable Audio-Visual Source Localizer (AAAI 2024 · similarity 0.86)
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.
