Visual Inertial Semantic Scene Representation for 3D Object Detection
We describe a system to detect objects in three-dimensional space using video and inertial sensors (accelerometer and gyrometer), ubiquitous in modern mobile platforms from phones to drones. Inertials afford the ability to impose class-specific scale priors for objects, and provide a global orientation reference. A minimal sufficient representation, the posterior of semantic (identity) and syntactic (pose) attributes of objects in space, can be decomposed into a geometric term, which can be maintained by a localization-and-mapping filter, and a likelihood function, which can be approximated by a discriminatively-trained convolutional neural network. The resulting system can process the video stream causally in real time, and provides a representation of objects in the scene that is persistent: Confidence in the presence of objects grows with evidence, and objects previously seen are kept in memory even when temporarily occluded, with their return into view automatically predicted to prime re-detection.
This work will be presented at CVPR 2017. The website is under construction. Please come back later. In the meantime, you can check out the supplementary videos below. An old version of the system has been used in a live demo at CVPR 2016 Demo session. The same system has been used in these videos.
- J. Dong, X. Fei and S. Soatto. Visual Inertial Semantic Scene Representation for 3D Object Detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. [pdf]