Banterle F., Gong R., Corsini M., Ganovelli F., Van Gool L., Cignoni P.
Time-frequency analysis Structure from motion Video Processing Video sequences Pipelines Computer architecture Streaming media Prediction algorithms Point-Cloud Generation Structure from Motion Deep Learning
Structure-from-Motion (SfM) using the frames of a video sequence can be a challenging task because there is a lot of redundant information, the computational time increases quadratically with the number of frames, there would be low-quality images (e.g., blurred frames) that can decrease the final quality of the reconstruction, etc. To overcome all these issues, we present a novel deep-learning architecture that is meant for speeding up SfM by selecting frames using predicted sub-sampling frequency. This architecture is general and can learn/distill the knowledge of any algorithm for selecting frames from a video for generating high-quality reconstructions. One key advantage is that we can run our architecture in real-time saving computations while keeping high-quality results.
Source: ICIP 2021 - 28th IEEE International Conference on Image Processing, pp. 3667–3671, Anchorage, Alaska, USA, 19-22/09/2021
Publisher: IEEE, New York, USA
@inproceedings{oai:it.cnr:prodotti:465907, title = {A deep learning method for frame selection in videos for structure from motion pipelines}, author = {Banterle F. and Gong R. and Corsini M. and Ganovelli F. and Van Gool L. and Cignoni P.}, publisher = {IEEE, New York, USA}, doi = {10.1109/icip42928.2021.9506227}, booktitle = {ICIP 2021 - 28th IEEE International Conference on Image Processing, pp. 3667–3671, Anchorage, Alaska, USA, 19-22/09/2021}, year = {2021} }