Low-Shot Visual Recognition by Shrinking and Hallucinating Features
Citations
1,897 citations
Cites background from "Low-Shot Visual Recognition by Shri..."
...ries included in the dataset, but blind, in principle, to other object categories outside the dataset, although ideally a powerful detection system should be able to recognize novel object categories [112,73]. Current detection datasets [53,179,129] contain only dozens to hundreds of categories, which is significantly smaller than those which can be recognized by humans. To achieve this goal, new large-sca...
[...]
1,082 citations
Cites background or methods from "Low-Shot Visual Recognition by Shri..."
...Logistic regression w/ H [4] (from [26]) 40....
[...]
...Table 3: Top-5 accuracy on the novel categories and on all categories (with and without priors) fot the ImageNet based few-shot benchmark proposed in [4] (for more details about the evaluation metrics we refer to [26])....
[...]
...We evaluate our framework on Mini-ImageNet and the recently introduced fews-shot benchmark of Bharath and Girshick [4] where we demonstrate that our approach is capable of both maintaining high recognition accuracy on base categories and to achieve excellent few-shot recognition accuracy on novel categories that surpasses prior state-of-the-art approaches by a significant margin....
[...]
...Finally, we apply our approach on the recently introduced few-shot benchmark of Bharath and Girshick [4] where we also achieve state-of-the-art results....
[...]
...Weight Gen entry) against prior work, such as Prototypical-Nets [22], Matching Networks [25], and the work of Bharath and Girshick [4]....
[...]
840 citations
Cites background from "Low-Shot Visual Recognition by Shri..."
...In Reference [53], by assuming that all categories share some transformable variability across samples, a single transformation function is learned to transfer variation between sample pairs learned from the other classes to (xi ,yi )....
[...]
831 citations
Cites background from "Low-Shot Visual Recognition by Shri..."
...This generalized few-shot setting is considered in an increasing number of studies [128], [165], [211]....
[...]
793 citations
References
123,388 citations
73,978 citations
49,639 citations
42,067 citations
8,289 citations
"Low-Shot Visual Recognition by Shri..." refers methods in this paper
...Amongst representation learning approaches, metric learning, such as the triplet loss [41, 38, 14] or siamese networks [22, 17], has been used to automatically learn feature representations where objects of the same class are closer together....
[...]