Radial Loss for Learning Fine-grained Video Similarity Metric
Citations
4 citations
References
73,978 citations
8,289 citations
7,091 citations
"Radial Loss for Learning Fine-grain..." refers methods in this paper
...Fine-grained Video Retrieval: (i) 3D-CNN [18], (ii) Triplet [19], (iii) Quadruplet-1 [4] (use loss function and quadruplet sampling strategy), (iv) Quadruplet-2(a,b) [20] (use loss function and two sampling strategies - where negative samples come from (a) both negative & intermediate categories, and (b) only from the negative class)....
[...]
3,786 citations
967 citations
"Radial Loss for Learning Fine-grain..." refers methods in this paper
...Fine-grained Video Retrieval: (i) 3D-CNN [18], (ii) Triplet [19], (iii) Quadruplet-1 [4] (use loss function and quadruplet sampling strategy), (iv) Quadruplet-2(a,b) [20] (use loss function and two sampling strategies - where negative samples come from (a) both negative & intermediate categories, and (b) only from the negative class)....
[...]
...Coarse grained video/clip similarity: (a) Triplet Precision (TP) relaxes the hard constraint of order preserving across 1Dataset Link: https://gofile.io/?c=8aphlD positive and intermediate samples....
[...]