Li Ying

TL;DR: In this paper, a cross-media search method using a VGG convolutional neural network (VGG net) to extract image features was proposed, where a Fisher vector based on Word2vec was utilized to extract text features and a correlation between the two heterogeneous features, which are images and text, was found by means of semantic matching based on logistic regression.

TL;DR: In this article, an image matching method and device is described, which includes the steps of: obtaining a template image and target images; obtaining a group of template characteristics according to the template image; extracting a group from the template images and extracting a target image according to target images, calculating image similarity of template image to each frame of target image, and determining the target image with the largest image similarity as the image matched with the template.

TL;DR: In this paper, a pedestrian detection method based on aggregated channel features is proposed, which comprises an early stage position calibration process and a later stage position screening process, in which multiple pieces of channel feature information in an input video or an image are aggregated, an image pyramid of the input image or the video in scale space is built and an image for each channel feature is calculated for each scale.

TL;DR: In this article, a deep discrimination network is constructed, different input images are fused and spliced in a color channel and the obtained splicing result is defined as the original difference space of different images.