Instance-aware Exploration-Verification-Exploitation for Instance ImageGoal Navigation
It addresses the problem of navigating to specific objects in unexplored environments for embodied AI agents, representing an incremental improvement over existing methods.
The paper tackles the Instance ImageGoal Navigation (IIN) task by proposing a modular framework that imitates human 'getting closer to confirm' behavior, achieving state-of-the-art success rates of 0.684 and 0.702 on the HM3D-SEM dataset compared to 0.561.
As a new embodied vision task, Instance ImageGoal Navigation (IIN) aims to navigate to a specified object depicted by a goal image in an unexplored environment. The main challenge of this task lies in identifying the target object from different viewpoints while rejecting similar distractors. Existing ImageGoal Navigation methods usually adopt the simple Exploration-Exploitation framework and ignore the identification of specific instance during navigation. In this work, we propose to imitate the human behaviour of ``getting closer to confirm" when distinguishing objects from a distance. Specifically, we design a new modular navigation framework named Instance-aware Exploration-Verification-Exploitation (IEVE) for instance-level image goal navigation. Our method allows for active switching among the exploration, verification, and exploitation actions, thereby facilitating the agent in making reasonable decisions under different situations. On the challenging HabitatMatterport 3D semantic (HM3D-SEM) dataset, our method surpasses previous state-of-the-art work, with a classical segmentation model (0.684 vs. 0.561 success) or a robust model (0.702 vs. 0.561 success)