Exploiting Scene-specific Features for Object Goal Navigation
This work addresses the challenge of efficient training for navigation models in robotics, though it is incremental as it builds on existing object navigation tasks.
The paper tackles the problem of object goal navigation in domestic environments by introducing a reduced dataset that speeds up training and an attention-based model (SMTSC) that exploits scene-object correlations, showing quantitative improvements in navigation performance.
Can the intrinsic relation between an object and the room in which it is usually located help agents in the Visual Navigation Task? We study this question in the context of Object Navigation, a problem in which an agent has to reach an object of a specific class while moving in a complex domestic environment. In this paper, we introduce a new reduced dataset that speeds up the training of navigation models, a notoriously complex task. Our proposed dataset permits the training of models that do not exploit online-built maps in reasonable times even without the use of huge computational resources. Therefore, this reduced dataset guarantees a significant benchmark and it can be used to identify promising models that could be then tried on bigger and more challenging datasets. Subsequently, we propose the SMTSC model, an attention-based model capable of exploiting the correlation between scenes and objects contained in them, highlighting quantitatively how the idea is correct.