CNN Features off-the-shelf: an Astounding Baseline for Recognition
Citations
55,235 citations
49,914 citations
14,635 citations
Cites methods from "CNN Features off-the-shelf: an Asto..."
...In fact, deep MPCNNs pre-trained by SL can extract useful features from quite diverse off-training-set images, yielding better results than traditional, widely used features such as SIFT (Lowe, 1999, 2004) on many vision tasks (Razavian et al., 2014)....
[...]
7,826 citations
5,919 citations
References
73,978 citations
40,826 citations
"CNN Features off-the-shelf: an Asto..." refers methods in this paper
...2 +C X i max(1 y iwT x i;0) (1) When training an SVM for one class we used all the images containing an instance of that class as the positive samples and the rest as negative samples. We used libsvm [9] with the trade-off parameter set to C=5 for all classes and chosen by cross-validation on the training set. Additionally, we augmented the set of positive samples by mirroring the positive images. We...
[...]
30,811 citations
21,729 citations
13,081 citations
"CNN Features off-the-shelf: an Asto..." refers background in this paper
...Most likely this question has been posed in your group’s coffee room....
[...]