Visualizing Object Detection Features
Links to Fileshttps://link.springer.com/article/10.1007/s11263-016-0884-7
MetadataShow full item record
Type of Work14 pages
journal articles preprints
Citation of Original PublicationVondrick, C., Khosla, A., Pirsiavash, H. et al. Int J Comput Vis (2016) 119: 145. https://doi.org/10.1007/s11263-016-0884-7
RightsThis item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
We introduce algorithms to visualize feature spaces used by object detectors. Our method works by inverting a visual feature back to multiple natural images. We found that these visualizations allow us to analyze object detection systems in new ways and gain new insight into the detector’s failures. For example, when we visualize the features for high scoring false alarms, we discovered that, although they are clearly wrong in image space, they often look deceptively similar to true positives in feature space. This result suggests that many of these false alarms are caused by our choice of feature space, and supports that creating a better learning algorithm or building bigger datasets is unlikely to correct these errors without improving the features. By visualizing feature spaces, we can gain a more intuitive understanding of recognition systems.