Tactical Rewind: Self-Correction via Backtracking in Vision-And-Language Navigation

We present the Frontier Aware Search with backTracking (FAST) Navigator, a general framework for action decoding, that achieves state-of-the-art results on the 2018 Room-to-Room (R2R) Vision-and-Language navigation challenge. Given a natural language instruction and photo-realistic image views of a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 6734 - 6742
Hauptverfasser: Ke, Liyiming, Li, Xiujun, Bisk, Yonatan, Holtzman, Ari, Gan, Zhe, Liu, Jingjing, Gao, Jianfeng, Choi, Yejin, Srinivasa, Siddhartha
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2019
Schlagworte:
Backtracking
Computer vision
Decoding
Natural languages
Navigation
Pattern recognition
Photorealistic images
Robotics + Driving
Trajectory
Vision + Language
Visualization
ISSN:1063-6919
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present the Frontier Aware Search with backTracking (FAST) Navigator, a general framework for action decoding, that achieves state-of-the-art results on the 2018 Room-to-Room (R2R) Vision-and-Language navigation challenge. Given a natural language instruction and photo-realistic image views of a previously unseen environment, the agent was tasked with navigating from source to target location as quickly as possible. While all current approaches make local action decisions or score entire trajectories using beam search, ours balances local and global signals when exploring an unobserved environment. Importantly, this lets us act greedily but use global signals to backtrack when necessary. Applying FAST framework to existing state-of-the-art models achieved a 17% relative gain, an absolute 6% gain on Success rate weighted by Path Length.
ISSN:1063-6919
DOI:10.1109/CVPR.2019.00690