HOME: Heatmap Output for future Motion Estimation

In this paper, we propose HOME, a framework tackling the motion forecasting problem with an image output representing the probability distribution of the agent's future location. This method allows for a simple architecture with classic convolution networks coupled with attention mechanism for...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2021 IEEE International Intelligent Transportation Systems Conference (ITSC) s. 500 - 507
Hlavní autoři: Gilles, Thomas, Sabatini, Stefano, Tsishkou, Dzmitry, Stanciulescu, Bogdan, Moutarde, Fabien
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 19.09.2021
Témata:
Benchmark testing
Heating systems
Motion estimation
Probability distribution
Sampling methods
Trajectory
Uncertainty
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this paper, we propose HOME, a framework tackling the motion forecasting problem with an image output representing the probability distribution of the agent's future location. This method allows for a simple architecture with classic convolution networks coupled with attention mechanism for agent interactions, and outputs an unconstrained 2D top-view representation of the agent's possible future. Based on this output, we design two methods to sample a finite set of agent's future locations. These methods allow us to control the optimization trade-off between miss rate and final displacement error for multiple modalities without having to retrain any part of the model. We apply our method to the Argoverse Motion Forecasting Benchmark and achieve 1 st place on the online leaderboard.
DOI:10.1109/ITSC48978.2021.9564944