Research map / Time series / Long-term forecasting
MX-LSTM: Mixing Tracklets and Vislets to Jointly Forecast Trajectories and Head Poses
Irtiza Hasan, Francesco Setti, Theodore Tsesmelis, Alessio Del Bue, Fabio Galasso, Marco Cristani
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 social pooling to the individual tracklet-and-vislet model (ablation) | MX-LSTM · Zara01, Zara02, UCY, TownCentre | MAD: 0.63 -> 0.59; 0.72 -> 0.35; 0.53 -> 0.49; 2.09 -> 1.15 | Improved |
| Add social pooling to the individual tracklet-and-vislet model (ablation) | MX-LSTM · Zara01, Zara02, UCY, TownCentre | FAD: 1.37 -> 1.31; 1.56 -> 0.79; 1.16 -> 1.12; 4.00 -> 2.30 | Improved |
| Use view-frustum social pooling instead of unrestricted social pooling (ablation) | MX-LSTM · Zara01, Zara02, UCY, TownCentre | MAD: 0.63 -> 0.59; 0.36 -> 0.35; 0.51 -> 0.49; 1.70 -> 1.15 | Improved |
| Use view-frustum social pooling instead of unrestricted social pooling (ablation) | MX-LSTM · Zara01, Zara02, UCY, TownCentre | FAD: 1.40 -> 1.31; 0.84 -> 0.79; 1.15 -> 1.12; 3.40 -> 2.30 | Improved |
| Add cross-stream covariance terms by using full rather than block-diagonal covariance (ablation) | MX-LSTM · Zara01, Zara02, UCY, TownCentre | MAD: 0.60 -> 0.59; 0.41 -> 0.35; 0.54 -> 0.49; 1.40 -> 1.15 | Mixed |
| Add cross-stream covariance terms by using full rather than block-diagonal covariance (ablation) | MX-LSTM · Zara01, Zara02, UCY, TownCentre | FAD: 1.51 -> 1.31; 1.00 -> 0.79; 1.23 -> 1.12; 2.90 -> 2.30 | Improved |
Closest papers by meaning
- Forecasting People Trajectories and Head Poses by Jointly Reasoning on Tracklets and Vislets (TPAMI 2019 · similarity 0.96)
- Social LSTM: Human Trajectory Prediction in Crowded Spaces (CVPR 2016 · similarity 0.88)
- A Unified Environmental Network for Pedestrian Trajectory Prediction (AAAI 2024 · similarity 0.86)
- Multi-modal Knowledge Distillation-based Human Trajectory Forecasting (CVPR 2025 · similarity 0.86)
- GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes (CVPR 2024 · similarity 0.86)
- Social-Transmotion: Promptable Human Trajectory Prediction (ICLR 2024 · similarity 0.85)
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.
