Research map / Label-efficient and robust learning / Recognition and long-tail

TSC-Net: Prediction of Pedestrian Trajectories by Trajectory-Scene-Cell Classification

ICLR 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes TSC-Net, a pedestrian trajectory prediction model that classifies trajectory-scene-cell relations to capture how scene semantics and goals shape future paths.

Problem

Predicting future pedestrian trajectories requires modeling scene information and how it relates to trajectories; existing methods using CNNs (and related approaches) are cut off in the available abstract.

Approach

Frames scene-aware trajectory prediction as trajectory-scene-cell classification, treating possible goal positions in the scene as cells to classify, so the model learns the relation between scene and trajectory.

Closest papers by meaning

• A Unified Environmental Network for Pedestrian Trajectory Prediction (AAAI 2024 · similarity 0.89)
• TrajCLIP: Pedestrian trajectory prediction method using contrastive learning and idempotent networks (NeurIPS 2024 · similarity 0.89)
• Towards Predicting Any Human Trajectory In Context (NeurIPS 2025 · similarity 0.87)
• 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 0.86)
• Pedestrian Trajectory Prediction with Missing Data: Datasets, Imputation, and Benchmarking (NeurIPS 2024 · similarity 0.85)