History Aware Multimodal Transformer for Vision-and-Language Navigation - Archive ouverte HAL Access content directly
Conference Papers Year :

History Aware Multimodal Transformer for Vision-and-Language Navigation

(1, 2) , (1, 2) , (1, 2) , (1, 2)
1
2

Abstract

Vision-and-language navigation (VLN) aims to build autonomous visual agents that follow instructions and navigate in real scenes. To remember previously visited locations and actions taken, most approaches to VLN implement memory using recurrent states. Instead, we introduce a History Aware Multimodal Transformer (HAMT) to incorporate a long-horizon history into multimodal decision making. HAMT efficiently encodes all the past panoramic observations via a hierarchical vision transformer (ViT), which first encodes individual images with ViT, then models spatial relation between images in a panoramic observation and finally takes into account temporal relation between panoramas in the history. It, then, jointly combines text, history and current observation to predict the next action. We first train HAMT end-to-end using several proxy tasks including single step action prediction and spatial relation prediction, and then use reinforcement learning to further improve the navigation policy. HAMT achieves new state of the art on a broad range of VLN tasks, including VLN with fine-grained instructions (R2R, RxR), high-level instructions (R2R-Last, REVERIE), dialogs (CVDN) as well as long-horizon VLN (R4R, R2R-Back). We demonstrate HAMT to be particularly effective for navigation tasks with longer trajectories.
Fichier principal
Vignette du fichier
hamt_paper.pdf (2.43 Mo) Télécharger le fichier
Vignette du fichier
hamt_supp.pdf (6.85 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03464975 , version 1 (03-12-2021)

Identifiers

  • HAL Id : hal-03464975 , version 1

Cite

Shizhe Chen, Pierre-Louis Guhur, Cordelia Schmid, Ivan Laptev. History Aware Multimodal Transformer for Vision-and-Language Navigation. NeurIPS 2021 - Thirty-fifth Conference on Neural Information Processing Systems, Dec 2021, Sydney / Virtual, Australia. ⟨hal-03464975⟩
45 View
103 Download

Share

Gmail Facebook Twitter LinkedIn More