No abstract

No abstract

No abstract

Recurrent neural network (RNN) is one of the most popular architectures for addressing variable sequence text, and it shows outstanding results in many natural language processing (NLP) tasks and remarkable performance in capturing long-term dependencies. Many models have achieved excellent results based on RNN. However, most of these models overlook the locations of the keywords in a sentence and the semantic connections in different directions. As a consequence, these methods do not make full use of the available information. Considering that different words in a sequence usually have different importance, in this paper, we propose bidirectional gated recurrent units (BGRUs) which integrates a novel attention pooling mechanism with max-pooling operation to force the model to pay attention to the keywords in a sentence and maintain the most meaningful information of the text automatically. The presented model allows to encode longer sequences. Thus, it not only prevents important information from being discarded but also can be used to filter noises. To avoid full exposure of content without any control, we add an output gate to the GRU, which is named as text unit. The proposed model was evaluated on multiple tasks, including sentimental classification, movie review data, and a subjective classification dataset. Experimental results show that our model can achieve excellent performance on these tasks.

No abstract

The 1st Tiny Object Detection (TOD) Challenge aims to encourage research in developing novel and accurate methods for tiny object detection in images which have wide views, with a current focus on tiny person detection. The TinyPerson dataset was used for the TOD Challenge and is publicly released. It has 1610 images and 72651 box-level annotations. Around 36 participating teams from the globe competed in the 1st TOD Challenge. In this paper, we provide a brief summary of the 1st TOD Challenge including brief introductions to the top three methods.The submission leaderboard will be reopened for researchers that are interested in the TOD challenge. The benchmark dataset and other information can be found at: https://github.com/ucas-vg/TinyBenchmark.

Estimating homography from an image pair is a fundamental problem in image alignment. Unsupervised learning methods have received increasing attention in this field due to their promising performance and label-free training. However, existing methods do not explicitly consider the problem of plane-induced parallax, which will make the predicted homography compromised on multiple planes. In this work, we propose a novel method HomoGAN to guide unsupervised homography estimation to focus on the dominant plane. First, a multi-scale transformer network is designed to predict homography from the feature pyramids of input images in a coarse-to-fine fashion. Moreover, we propose an unsupervised GAN to impose coplanarity constraint on the predicted homography, which is realized by using a generator to predict a mask of aligned regions, and then a discriminator to check if two masked feature maps are induced by a single homography. To validate the effectiveness of HomoGAN and its components, we conduct extensive experiments on a large-scale dataset, and results show that our matching error is 22% lower than the previous SOTA method. Code is available at https: //github.com/megvii-research/HomoGAN .

No abstract