SCOD: Active Object Detection for Embodied Agents using Sensory Commutativity of Action Sequences
This addresses the challenge of object discovery for naive embodied agents, offering a novel approach that generalizes to unseen environments and objects, though it appears incremental in the context of active detection methods.
The paper tackles the problem of object detection for embodied agents by introducing SCOD, a method that exploits the commutative properties of action sequences to compare observations from different action orders, achieving accurate results in 3D robotic simulations and real-world applications.
We introduce SCOD (Sensory Commutativity Object Detection), an active method for movable and immovable object detection. SCOD exploits the commutative properties of action sequences, in the scenario of an embodied agent equipped with first-person sensors and a continuous motor space with multiple degrees of freedom. SCOD is based on playing an action sequence in two different orders from the same starting point and comparing the two final observations obtained after each sequence. Our experiments on 3D realistic robotic setups (iGibson) demonstrate the accuracy of SCOD and its generalization to unseen environments and objects. We also successfully apply SCOD on a real robot to further illustrate its generalization properties. With SCOD, we aim at providing a novel way of approaching the problem of object discovery in the context of a naive embodied agent. We provide code and a supplementary video.