Disentangled Acoustic Fields For Multimodal Physical Property Understanding
Disentangled Acoustic Fields For Multimodal Physical Property Understanding
This video provides a comprehensive view of the agent's perspective, comparing the performance of both the strong baseline approach (modular planning) and our proposed DAF in two distinct environments: the study room and kitchen. Additionally, the figure below showcases the trajectories taken by both methods and the corresponding uncertainty map in these scenarios. The demonstration effectively highlights the improved efficiency of our method in locating fallen objects, with a particular emphasis on the valuable insights offered by the uncertainty map.
We compare the agent trajectories using our method (Red) against the trajectories produced by the modular planning baseline (Green). The loss map uses dark blue to indicate regions of low error, while yellow is used to indicate regions of high error. This figure compares the uncertainty maps of various cases. Darker colors indicate lower values of position loss. The star (Gold) symbolizes the ground truth position of the fallen object. The end of each trajectory is circled in white for clarity. In (a)−(f), the baseline method fails to find the target, while our method succeeds. In (g)−(h), both methods find the target, but our method takes a shorter path.