HNIP: Compact Deep Invariant Representations for Video Matching, Localization, and Retrieval.
Jie LinLing-Yu DuanShiqi WangYan BaiYihang LouVijay ChandrasekharTiejun HuangAlex ChiChung KotWen GaoPublished in: IEEE Trans. Multim. (2017)
Keyphrases
- invariant representations
- invariant representation
- video sequences
- object recognition
- multimedia
- video data
- algebraic curves
- matching algorithm
- video content
- video retrieval
- feature points
- invariant properties
- recognizing objects
- video frames
- image retrieval
- image matching
- key frames
- three dimensional
- space time
- keypoints
- multiscale