CCLSTM: Coupled Convolutional Long-Short Term Memory Network for Occupancy Flow Forecasting

2025/06/06 by Peter Lengyel, Lengyel, Peter
Engineering · Social Sciences · #Traffic Prediction and Management Techniques #Autonomous Vehicle Technology and Safety #Human Mobility and Location-Based Analysis

paper · pdf · doi:10.48550/arxiv.2506.06128

Abstract

Predicting future states of dynamic agents is a fundamental task in autonomous driving. An expressive representation for this purpose is Occupancy Flow Fields, which provide a scalable and unified format for modeling motion, spatial extent, and multi-modal future distributions. While recent methods have achieved strong results using this representation, they often depend on high-quality vectorized inputs, which are unavailable or difficult to generate in practice, and the use of transformer-based architectures, which are computationally intensive and costly to deploy. To address these issues, we propose Coupled Convolutional LSTM (CCLSTM), a lightweight, end-to-end trainable architecture based solely on convolutional operations. Without relying on vectorized inputs or self-attention mechanisms, CCLSTM effectively captures temporal dynamics and spatial occupancy-flow correlations using a compact recurrent convolutional structure. Despite its simplicity, CCLSTM achieves state-of-the-art performance on occupancy flow metrics and, as of this submission, ranks \(1^st\) in all metrics on the 2024 Waymo Occupancy and Flow Prediction Challenge leaderboard.

Citations

• HGNET: A Hierarchical Feature Guided Network for Occupancy Flow Field Prediction
• OFMPNet: Deep End-to-End Model for Occupancy and Flow Prediction in Urban Environment
• Hybrid-Prediction Integrated Planning for Autonomous Driving
• Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting
• VectorFlow: Combining Images and Vectors for Traffic Occupancy and Flow Prediction
• HOPE: Hierarchical Spatial-temporal Network for Occupancy Flow Prediction
• Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?
• Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Dataset
• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
• Unsupervised Learning of Video Representations using LSTMs
• Sequence to Sequence Learning with Neural Networks
• Long Short-Term Memory

Related

• HGNET: A Hierarchical Feature Guided Network for Occupancy Flow Field Prediction
• VectorFlow: Combining Images and Vectors for Traffic Occupancy and Flow Prediction
• HOPE: Hierarchical Spatial-temporal Network for Occupancy Flow Prediction
• OFMPNet: Deep End-to-End Model for Occupancy and Flow Prediction in Urban Environment
• Double-Prong ConvLSTM for Spatiotemporal Occupancy Prediction in Dynamic Environments