🧪 Test?View on arXiv
MESSENGER: Memory-Enhanced Sequential Scene Flow Estimation via Autoregressive Next-Frame Forecasting
Liu Jiuming, Author 2, Author 3, Author 4, Author 5
scene flowautoregressive forecastingmemory networkstemporal dependencies
2610.10759
Builder Relevance
2h ago80%
Abstract
The paper presents MESSENGER, a memory-enhanced pipeline for scene flow estimation that improves long-term temporal correlation and reduces prediction errors.
Reality Card
Core Claim
MESSENGER achieves state-of-the-art performance in long-horizon future extrapolation, reducing EPE3D by 71.6% on nuScenes and 67.7% on Argoverse 2.
Method / Result
Reduces EPE3D by 71.6% on nuScenes.
Limitations
The computational overhead may increase with more input frames, potentially affecting reproducibility in resource-constrained environments.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.