Context-aware deep feature compression for high-speed visual tracking
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
Colleges, School and Institutes
- Clova AI Research, NAVER Corp.
- Seoul National University
- Imperial College London
We propose a new context-aware correlation ﬁlter based tracking framework to achieve both high computational speed and state-of-the-art performance among real-time trackers. The major contribution to the high computational speed lies in the proposed deep feature compression that is achieved by a context-aware scheme utilizing multiple expert auto-encoders; a context in our framework refers to the coarse category of the tracking target according to appearance patterns. In the pre-training phase, one expert auto-encoder is trained per category. In the tracking phase, the best expert auto-encoder is selected for a given target, and only this auto-encoder is used. To achieve high tracking performance with the compressed feature map, we introduce extrinsic denoising processes and a new orthogonality loss term for pre-training and ﬁne-tuning of the expert autoencoders. We validate the proposed context-aware framework through a number of experiments, where our method achieves a comparable performance to state-of-the-art trackers which cannot run in real-time, while running at a signiﬁcantly fast speed of over 100 fps.
|Title of host publication||Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018)|
|Publication status||Published - 17 Dec 2018|
|Event||IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018) - Salt Lake City, United States|
Duration: 18 Jun 2018 → 22 Jun 2018
|Name||Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition)|
|Conference||IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2018)|
|City||Salt Lake City|
|Period||18/06/18 → 22/06/18|