Towards universal representation for unseen action recognition
Abstract Unseen Action Recognition (UAR) aims to recognise novel action categories
without training examples. While previous methods focus on inner-dataset seen/unseen …
without training examples. While previous methods focus on inner-dataset seen/unseen …
Mining mid-level features for image classification
Mid-level or semi-local features learnt using class-level information are potentially more
distinctive than the traditional low-level local features constructed in a purely bottom-up …
distinctive than the traditional low-level local features constructed in a purely bottom-up …
Face recognition with a small occluded training set using spatial and statistical pooling
Occlusion is one of the most intractable problems for face recognition. Double-occlusion
problem is an extremely challenging case that the occlusion can occur in both of training …
problem is an extremely challenging case that the occlusion can occur in both of training …
When naive bayes nearest neighbors meet convolutional neural networks
I Kuzborskij, FM Carlucci… - Proceedings of the IEEE …, 2016 - openaccess.thecvf.com
Abstract Since Convolutional Neural Networks (CNNs) have become the leading learning
paradigm in visual recognition, Naive Bayes Nearest Neighbor (NBNN)-based classifiers …
paradigm in visual recognition, Naive Bayes Nearest Neighbor (NBNN)-based classifiers …
Supervised local descriptor learning for human action recognition
Local features have been widely used in computer vision tasks, eg, human action
recognition, but it tends to be an extremely challenging task to deal with large-scale local …
recognition, but it tends to be an extremely challenging task to deal with large-scale local …
Cross-season place recognition using nbnn scene descriptor
T Kanji - 2015 IEEE/RSJ International Conference on Intelligent …, 2015 - ieeexplore.ieee.org
We propose a discriminative compact scene descriptor for single-view cross-season place
recognition. Unlike previous bag-of-words approaches which rely on a library of vector …
recognition. Unlike previous bag-of-words approaches which rely on a library of vector …
基于朴素贝叶斯K近邻的快速图像分类算法
张旭, 蒋建国, 洪日昌, 杜跃 - 北京航空航天大学学报, 2015 - bhxb.buaa.edu.cn
朴素贝叶斯最近邻(NBNN) 分类算法具有非特征量化和图像-类别度量方式的优点,
但算法运行速度较慢, 分类正确率较低. 针对此问题, 提出一种朴素贝叶斯K 近邻分类算法 …
但算法运行速度较慢, 分类正确率较低. 针对此问题, 提出一种朴素贝叶斯K 近邻分类算法 …
[PDF][PDF] Discriminative Embedding via Image-to-Class Distances.
Abstract Image-to-Class (I2C) distance firstly proposed in the naive Bayes nearest
neighbour (NBNN) classifier has shown its effectiveness in image classification. However …
neighbour (NBNN) classifier has shown its effectiveness in image classification. However …
Local coding based matching kernel method for image classification
Y Song, IV McLoughlin, LR Dai - PloS one, 2014 - journals.plos.org
This paper mainly focuses on how to effectively and efficiently measure visual similarity for
local feature based representation. Among existing methods, metrics based on Bag of Visual …
local feature based representation. Among existing methods, metrics based on Bag of Visual …
Mining DCNN landmarks for long-term visual SLAM
T Taisho, T Kanji - 2016 IEEE International Conference on …, 2016 - ieeexplore.ieee.org
Long-term visual SLAM, in familiar, semi-dynamic, and partially changing environments is
an important area of research in robotics. The main problem we faced is the question of how …
an important area of research in robotics. The main problem we faced is the question of how …