[Sep 2015 ~ Oct 2015] Deep Learning in arxiv

Scale-awareFast R-CNN for Pedestrian Detection

AggregatingDeep Convolutional Features for Image Retrieval

这种特征aggregate的方式，在几个数据集上有一定增长

LearningMulti-Domain Convolutional Neural Networks for Visual Tracking

multitask在视觉跟踪中的应用

VideoParagraph Captioning using Hierarchical Recurrent Neural Networks

度长在视频段落描述方面的一个力作

Attention with Intention for a Neural Network Conversation Model

EmpiricalStudy on Deep Learning Models for QA

IBM Watson