Research map / Detection and segmentation / Tracking
Deep Tracking: Seeing Beyond Seeing 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 |
|---|---|---|---|
| train with unsupervised sensor-only data instead of supervised ground-truth scene states | Deep Tracking · Synthetic 2D tracking | tracking performance | No clear effect |
| use future prediction with spatial and temporal input dropout to train from visible observations | Deep Tracking · Synthetic 2D tracking | prediction of occluded objects | Helped |
| change object shapes from circles to squares | Deep Tracking · Synthetic 2D tracking with square objects | world-state prediction | No clear effect |
| add sensor noise to 1% of pixel observations | Deep Tracking · Synthetic 2D tracking with sensor noise | world-state prediction | No clear effect |
Closest papers by meaning
- Online Multi-Target Tracking Using Recurrent Neural Networks (AAAI 2017 · similarity 0.87)
- Backprop KF: Learning Discriminative Deterministic State Estimators (NeurIPS 2016 · similarity 0.87)
- Tracking Without Re-recognition in Humans and Machines (NeurIPS 2021 · similarity 0.87)
- Depth From Camera Motion and Object Detection (CVPR 2021 · similarity 0.86)
- SUTrack: Towards Simple and Unified Single Object Tracking (AAAI 2025 · similarity 0.85)
- Track-On: Transformer-based Online Point Tracking with Memory (ICLR 2025 · similarity 0.84)
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.