1 code implementation • CVPR 2023 • Takumi Kobayashi
In contrast to the softmax loss, the BCE loss involves issues regarding imbalance as multiple classes are decomposed into a bunch of binary classifications; recent works improve the BCE loss to cope with the issue by means of weighting.
no code implementations • 8 Nov 2021 • Lincon S. Souza, Naoya Sogi, Bernardo B. Gatto, Takumi Kobayashi, Kazuhiro Fukui
The image set is represented by a low-dimensional input subspace; and this input subspace is matched with reference subspaces by a similarity of their canonical angles, an interpretable and easy to compute metric.
1 code implementation • CVPR 2021 • Takumi Kobayashi
By focusing on the angle between a feature and a classifier which is embedded in cosine similarity at the classification layer, we formulate a novel similarity beyond the cosine based on von Mises-Fisher distribution of directional statistics.
1 code implementation • NeurIPS 2019 • Takumi Kobayashi
Convolutional neural networks (CNNs) contain local pooling to effectively downsize feature maps for increasing computation efficiency as well as robustness to input variations.
no code implementations • 29 Oct 2019 • Kazuhiro Fukui, Naoya Sogi, Takumi Kobayashi, Jing-Hao Xue, Atsuto Maki
To avoid the difficulty, we first introduce geometrical Fisher discriminant analysis (gFDA) based on a simplified Fisher criterion.
no code implementations • ICCV 2019 • Takumi Kobayashi
In the proposed method, the parameterized pooling form is derived from a probabilistic perspective to flexibly represent various types of pooling and then the parameters are estimated by means of global statistics in the input feature map.
no code implementations • CVPR 2018 • Takumi Kobayashi
Deep convolutional neural network (ConvNet) is a promising approach for high-performance image classification.
no code implementations • ICCV 2017 • Takumi Kobayashi
In action recognition, local motion descriptors contribute to effectively representing video sequences where target actions appear in localized spatio-temporal regions.
no code implementations • CVPR 2016 • Takumi Kobayashi
Unlike the previous methods, the proposed method is built on not a histogram form but a tensor structure of a feature array extracted such as on spatial grids, in order to construct effective SSIM-based similarity measure of high robustness which is a key requirement in feature matching.
no code implementations • CVPR 2015 • Takumi Kobayashi
In contrast to category-level or cluster-level classifiers, exemplar SVM is successfully applied to classifying (or detecting) a target object as well as transferring instance-level annotations.
no code implementations • CVPR 2014 • Takumi Kobayashi
On the other hand, in the bag-of-feature (BoF) framework, the Dirichlet mixture model can be extended to Gaussian mixture by transforming histogram-based local descriptors, e. g., SIFT, and thereby we propose the method of Dirichlet-derived GMM Fisher kernel.
no code implementations • CVPR 2013 • Takumi Kobayashi
Since the p. d. f essentially represents the image, we extract the features from the p. d. f by means of the gradients on the p. d. f.