Binaural SoundNet: Predicting Semantics, Depth and Motion with Binaural Sounds

This work develops an approach for scene understanding purely based on binaural sounds. The considered tasks include predicting the semantic masks of sound-making objects, the motion of sound-making objects, and the depth map of the scene. To this aim, we propose a novel sensor setup and record a new audio-visual dataset of street scenes with eight professional binaural microphones and a 360camera. The co-existence of visual and audio cues is leveraged for supervision transfer. In particular, we employ a cross-modal distillation framework that consists of multiple vision teacher methods and a sound student method the student method is trained to generate the same results as the teacher methods do. This way, the auditory system can be trained without using human annotations. To further boost the performance, we propose another novel auxiliary task, coined Spatial Sound Super- Resolution, to increase the directional resolution of sounds. We then formulate the four tasks into one end-to-end trainable multi-tasking network aiming to boost the overall performance. Experimental results show that 1) our method achieves good results for all four tasks, 2) the four tasks are mutually beneficial, and 3) the number and orientation of microphones are both importantant.