The company also publishes an arxiv.org paper "Speed Invariant Time Surface for Learning to Detect Corner Points with Event-Based Cameras" by Jacques Manderscheid, Amos Sironi, Nicolas Bourdis, Davide Migliore, and Vincent Lepetit.
"We propose a learning approach to corner detection for event-based cameras that is stable even under fast and abrupt motions. Event-based cameras offer high temporal resolution, power efficiency, and high dynamic range. However, the properties of event-based data are very different compared to standard intensity images, and simple extensions of corner detection methods designed for these images do not perform well on event-based data. We first introduce an efficient way to compute a time surface that is invariant to the speed of the objects. We then show that we can train a Random Forest to recognize events generated by a moving corner from our time surface. Random Forests are also extremely efficient, and therefore a good choice to deal with the high capture frequency of event-based cameras ---our implementation processes up to 1.6Mev/s on a single CPU. Thanks to our time surface formulation and this learning approach, our method is significantly more robust to abrupt changes of direction of the corners compared to previous ones. Our method also naturally assigns a confidence score for the corners, which can be useful for postprocessing. Moreover, we introduce a high-resolution dataset suitable for quantitative evaluation and comparison of corner detection methods for event-based cameras. We call our approach SILC, for Speed Invariant Learned Corners, and compare it to the state-of-the-art with extensive experiments, showing better performance."
Thanks to TL for the pointer!
Such kind of visual features can be extracted from a classic image. Some commercial automatic transportation vehicules already use such corner points to mapping and navigation. What could be the fundamental advantages of event based approach if complexe post-processing is needed ?
ReplyDeleteSuch kind of features could be extracted from images of conventional cameras but ...
ReplyDelete* it costs some processing resources to extract such features from images of conventional cameras
* conventional cameras captures lots of redundant data from static backgrounds
* not many conventional cameras allow to get data at >10kfps :) and even if exist they cost a lot
* frames are captured at certain fps meaning that there is some gap between such frames, however Prophesee's sensors allows to get data continuously, without gaps