Search

As a fundamental problem in image understanding, image caption generation has attracted much attention from both computer vision and natural language processing communities. In this paper, we focus on how to exploit the structure information of a natural sentence, which is used to describe the content of an image. We discover that the Part of Speech (PoS) tags of a sentence, are very effective cues for guiding the Long Short-Term Memory (LSTM) based word generator. More specifically, given a sentence, the PoS tag of each word is utilized to determine whether it is essential to input image representation into the word generator. Benefiting from such a strategy, our model can closely connect the visual attributes of an image to the word concepts in the natural language space. Experimental results on the most popular benchmark datasets, e.g., Flickr30k and MS COCO, consistently demonstrate that our method can significantly enhance the performance of a standard image caption generation model, and achieve the conpetitive results.

Recently, attribute has demonstrated its effectiveness in guiding image captioning system. However, most attributes based image captioning methods treat the attributes prediction task as a separate task and rely on a standalone stage to obtain the attributes for the given image, e.g., a pre-trained network like Fully Convolutional Neural Network (FCN) is usually adopted. Inherently, they ignore the correlation between the attribute prediction task and image representation extraction task, and at the same time increases the complexity of the image captioning system. In this paper, we aim to couple the attributes prediction stage and image representation extraction stage tightly and propose a novel and efficient image captioning framework called Visual-Densely Semantic Attention Network(VD-SAN). In particular, the whole captioning system consists of shared convolutional layers from Dense Convolutional Network (DenseNet), which are further split into a semantic attributes prediction branch and an image feature extraction branch, two semantic attention models, and a long short-term memory networks (LSTM) for caption generation. To evaluate the proposed architecture, we construct Flickr30K-ATT and MS-COCO-ATT datasets based on the original popular image caption datasets Flickr30K and MS COCO respectively, and each image from Flickr30K-ATT or MS-COCO-ATT is annotated with an attribute list in addition to the corresponding caption. Empirical results demonstrate that our captioning system can achieve significant improvements over state-of-the-art approaches.

This paper proposes a new generative adversarial network for pose transfer, i.e., transferring the pose of a given person to a target pose. We design a progressive generator which comprises a sequence of transfer blocks. Each block performs an intermediate transfer step by modeling the relationship between the condition and the target poses with attention mechanism. Two types of blocks are introduced, namely Pose-Attentional Transfer Block (PATB) and Aligned Pose-Attentional Transfer Bloc ~(APATB). Compared with previous works, our model generates more photorealistic person images that retain better appearance consistency and shape consistency compared with input images. We verify the efficacy of the model on the Market-1501 and DeepFashion datasets, using quantitative and qualitative measures. Furthermore, we show that our method can be used for data augmentation for the person re-identification task, alleviating the issue of data insufficiency. Code and pretrained models are available at https://github.com/tengteng95/Pose-Transfer.git., Accepted by TPAMI. An extension of the conference version. Conference version at arXiv:1904.03349

Scene text detection and recognition is a universal text recognition technology, which has become a hot research topic in the field of computer vision and document analysis in recent years. It is widely applied in geographical positioning, license plate recognition, and driverless applications. Compared to traditional document text detection and recognition, scene text varies more dramatically in font, color, scale, layout, and background. Owing to its excellent performance, deep learning has been widely adopted in this field. In this paper, we mainly review our representative studies based on deep learning in this field and describe the future research trends in this field.

Text in natural images is an important source of information, which can be utilized for many real-world applications. This work focuses on a new problem: distinguishing images that contain text from a large volume of natural images. To address this problem, we propose a novel convolutional neural network variant, called multi-scale spatial partition network (MSP-Net). The network classifies images that contain text or not, by predicting text existence in all image blocks, which are spatial partitions at multiple scales on an input image. The whole image is classified as a text image (an image containing text) as long as one of the blocks is predicted to contain text. The network classifies images very efficiently by predicting all blocks simultaneously in a single forward propagation. Through experimental evaluations and comparisons on public datasets, we demonstrate the effectiveness and robustness of the proposed method. HighlightsWe study a new and important problem: text/non-text image classification in the wild.A new scheme based on block-level classification is proposed to tackle this problem.We propose MSP-Net, a novel CNN variant, to efficiently classify text/non-text images.As a by-product, MSP-Net outputs coarse locations and scales of texts.

Chinese scene text reading is one of the most challenging problems in computer vision and has attracted great interest. Different from English text, Chinese has more than 6000 commonly used characters and Chinese characters can be arranged in various layouts with numerous fonts. The Chinese signboards in street view are a good choice for Chinese scene text images since they have different backgrounds, fonts and layouts. We organized a competition called ICDAR2019-ReCTS, which mainly focuses on reading Chinese text on signboard. This report presents the final results of the competition. A large-scale dataset of 25,000 annotated signboard images, in which all the text lines and characters are annotated with locations and transcriptions, were released. Four tasks, namely character recognition, text line recognition, text line detection and end-to-end recognition were set up. Besides, considering the Chinese text ambiguity issue, we proposed a multi ground truth (multi-GT) evaluation method to make evaluation fairer. The competition started on March 1, 2019 and ended on April 30, 2019. 262 submissions from 46 teams are received. Most of the participants come from universities, research institutes, and tech companies in China. There are also some participants from the United States, Australia, Singapore, and Korea. 21 teams submit results for Task 1, 23 teams submit results for Task 2, 24 teams submit results for Task 3, and 13 teams submit results for Task 4. The official website for the competition is http://rrc.cvc.uab.es/?ch=12.

This paper proposes a new generative adversarial network for pose transfer, i.e., transferring the pose of a given person to a target pose. The generator of the network comprises a sequence of Pose-Attentional Transfer Blocks that each transfers certain regions it attends to, generating the person image progressively. Compared with those in previous works, our generated person images possess better appearance consistency and shape consistency with the input images, thus significantly more realistic-looking. The efficacy and efficiency of the proposed network are validated both qualitatively and quantitatively on Market-1501 and DeepFashion. Furthermore, the proposed architecture can generate training images for person re-identification, alleviating data insufficiency. Codes and models are available at: https://github.com/tengteng95/Pose-Transfer.git., Comment: To appear in CVPR 2019, oral presentation (21 pages, 15 figures including the supplementary materials)

Script identification facilitates many important applications in document/video analysis. This paper investigates a relatively new problem: identifying scripts in natural images. The basic idea is combining deep features and mid-level representations into a globally trainable deep model. Specifically, a set of deep feature maps is firstly extracted by a pre-trained CNN model from the input images, where the local deep features are densely collected. Then, discriminative clustering is performed to learn a set of discriminative patterns based on such local features. A mid-level representation is obtained by encoding the local features based on the learned discriminative patterns (codebook). Finally, the mid-level representations and the deep features are jointly optimized in a deep network. Benefiting from such a fine-grained classification strategy, the optimized deep model, termed Discriminative Convolutional Neural Network (DisCNN), is capable of effectively revealing the subtle differences among the scripts difficult to be distinguished, e.g. Chinese and Japanese. In addition, a large scale dataset containing 16,291 in-the-wild text images in 13 scripts, namely SIW-13, is created for evaluation. Our method is not limited to identifying text images, and performs effectively on video and document scripts as well, not requiring any preprocess like binarization, segmentation or hand-crafted features. The experimental comparisons on the datasets including SIW-13, CVSI-2015 and Multi-Script consistently demonstrate DisCNN a state-of-the-art approach for script identification. HighlightsWe study a new and important topic: script identification in scene text images.The proposed DiscCNN combines deep features and the mid-level representation.DiscCNN learns special characteristics of scripts from training data automatically.DiscCNN achieves state-of-the-art performances on scene, video and document scripts.A large-scale in-the-wild script identification dataset is proposed.

This letter introduces a robust representation of 3-D shapes, named DeepPano, learned with deep convolutional neural networks (CNN). Firstly, each 3-D shape is converted into a panoramic view, namely a cylinder projection around its principle axis. Then, a variant of CNN is specifically designed for learning the deep representations directly from such views. Different from typical CNN, a row-wise max-pooling layer is inserted between the convolution and fully-connected layers, making the learned representations invariant to the rotation around a principle axis. Our approach achieves state-of-the-art retrieval/classification results on two large-scale 3-D model datasets (ModelNet-10 and ModelNet-40), outperforming typical methods by a large margin.

Despite the great success of convolutional neural networks (CNNs) for the image classification task on data sets such as Cifar and ImageNet, CNN’s representation power is still somewhat limited in dealing with images that have a large variation in size and clutter, where Fisher vector (FV) has shown to be an effective encoding strategy. FV encodes an image by aggregating local descriptors with a universal generative Gaussian mixture model (GMM). FV, however, has limited learning capability and its parameters are mostly fixed after constructing the codebook. To combine together the best of the two worlds, we propose in this brief a neural network structure with FV layer being part of an end-to-end trainable system that is differentiable; we name our network FisherNet that is learnable using back propagation. Our proposed FisherNet combines CNN training and FV encoding in a single end-to-end structure. We observe a clear advantage of FisherNet over plain CNN and standard FV in terms of both classification accuracy and computational efficiency on the challenging PASCAL visual object classes object classification and emotion image classification tasks.

Scene text detection is an important step of scene text recognition system and also a challenging problem. Different from general object detection, the main challenges of scene text detection lie on arbitrary orientations, small sizes, and significantly variant aspect ratios of text in natural images. In this paper, we present an end-to-end trainable fast scene text detector, named TextBoxes++, which detects arbitrary-oriented scene text with both high accuracy and efficiency in a single network forward pass. No post-processing other than an efficient non-maximum suppression is involved. We have evaluated the proposed TextBoxes++ on four public datasets. In all experiments, TextBoxes++ outperforms competing methods in terms of text localization accuracy and runtime. More specifically, TextBoxes++ achieves an f-measure of 0.817 at 11.6fps for 1024*1024 ICDAR 2015 Incidental text images, and an f-measure of 0.5591 at 19.8fps for 768*768 COCO-Text images. Furthermore, combined with a text recognizer, TextBoxes++ significantly outperforms the state-of-the-art approaches for word spotting and end-to-end text recognition tasks on popular benchmarks. Code is available at: https://github.com/MhLiao/TextBoxes_plusplus, Comment: 15 pages

A challenging aspect of scene text recognition is to handle text with distortions or irregular layout. In particular, perspective text and curved text are common in natural scenes and are difficult to recognize. In this work, we introduce ASTER, an end-to-end neural network model that comprises a rectification network and a recognition network. The rectification network adaptively transforms an input image into a new one, rectifying the text in it. It is powered by a flexible Thin-Plate Spline transformation which handles a variety of text irregularities and is trained without human annotations. The recognition network is an attentional sequence-to-sequence model that predicts a character sequence directly from the rectified image. The whole model is trained end to end, requiring only images and their groundtruth text. Through extensive experiments, we verify the effectiveness of the rectification and demonstrate the state-of-the-art recognition performance of ASTER. Furthermore, we demonstrate that ASTER is a powerful component in end-to-end recognition systems, for its ability to enhance the detector.

A significant challenge in scene text detection is the large variation in text sizes. In particular, small text are usually hard to detect. This paper presents an accurate oriented text detector based on Faster R-CNN. We observe that Faster R-CNN is suitable for general object detection but inadequate for scene text detection due to the large variation in text size. We apply feature fusion both in RPN and Fast R-CNN to alleviate this problem and furthermore, enhance model's ability to detect relatively small text. Our text detector achieves comparable results to those state of the art methods on ICDAR 2015 and MSRA-TD500, showing its advantage and applicability.

Text in natural images is of arbitrary orientations, requiring detection in terms of oriented bounding boxes. Normally, a multi-oriented text detector often involves two key tasks: 1) text presence detection, which is a classification problem disregarding text orientation; 2) oriented bounding box regression, which concerns about text orientation. Previous methods rely on shared features for both tasks, resulting in degraded performance due to the incompatibility of the two tasks. To address this issue, we propose to perform classification and regression on features of different characteristics, extracted by two network branches of different designs. Concretely, the regression branch extracts rotation-sensitive features by actively rotating the convolutional filters, while the classification branch extracts rotation-invariant features by pooling the rotation-sensitive features. The proposed method named Rotation-sensitive Regression Detector (RRD) achieves state-of-the-art performance on three oriented scene text benchmark datasets, including ICDAR 2015, MSRA-TD500, RCTW-17 and COCO-Text. Furthermore, RRD achieves a significant improvement on a ship collection dataset, demonstrating its generality on oriented object detection., accepted by CVPR 2018

This report presents the final results of the ICDAR 2017 Robust Reading Challenge on COCO-Text. A challenge on scene text detection and recognition based on the largest real scene text dataset currently available: the COCO-Text dataset. The competition is structured around three tasks: Text Localization, Cropped Word Recognition and End-To-End Recognition. The competition received a total of 27 submissions over the different opened tasks. This report describes the datasets and the ground truth, details the performance evaluation protocols used and presents the final results along with a brief summary of the participating methods.

Chinese is the most widely used language in the world. Algorithms that read Chinese text in natural images facilitate applications of various kinds. Despite the large potential value, datasets and competitions in the past primarily focus on English, which bares very different characteristics than Chinese. This report introduces RCTW, a new competition that focuses on Chinese text reading. The competition features a large-scale dataset with 12,263 annotated images. Two tasks, namely text localization and end-to-end recognition, are set up. The competition took place from January 20 to May 31, 2017. 23 valid submissions were received from 19 teams. This report includes dataset description, task definitions, evaluation protocols, and results summaries and analysis. Through this competition, we call for more future research on the Chinese text reading problem. The official website for the competition is http://rctw.vlrlab.net

Most state-of-the-art text detection methods are specific to horizontal Latin text and are not fast enough for real-time applications. We introduce Segment Linking (SegLink), an oriented text detection method. The main idea is to decompose text into two locally detectable elements, namely segments and links. A segment is an oriented box covering a part of a word or text line; A link connects two adjacent segments, indicating that they belong to the same word or text line. Both elements are detected densely at multiple scales by an end-to-end trained, fully-convolutional neural network. Final detections are produced by combining segments connected by links. Compared with previous methods, SegLink improves along the dimensions of accuracy, speed, and ease of training. It achieves an f-measure of 75.0% on the standard ICDAR 2015 Incidental (Challenge 4) benchmark, outperforming the previous best by a large margin. It runs at over 20 FPS on 512x512 images. Moreover, without modification, SegLink is able to detect long lines of non-Latin text, such as Chinese., Comment: To Appear in CVPR 2017

In this paper, we focus on the text/non-text classification problem: distinguishing images that contain text from a lot of natural images. To this end, we propose a novel neural network architecture, termed Convolutional Multi-Dimensional Recurrent Neural Network (CMDRNN), which distinguishes text/non-text images by classifying local image blocks, taking both region pixels and dependencies among blocks into account. The network is composed of a Convolutional Neural Network (CNN) and a Multi-Dimensional Recurrent Neural Network (MDRNN). The CNN extracts rich and high-level image representation, while the MDRNN analyzes dependencies along multiple directions and produces block-level predictions. By evaluating CMDRNN on a public dataset, we observe improvements over prior arts in terms of both speed and accuracy.

Script identification for scene text images is a challenging task. This paper describes a novel deep neural network structure that efficiently identifies scripts of images. In our design, we exploit two important factors, namely the image representation, and the spatial dependencies within text lines. To this end, we bring together a Convolutional Neural Network (CNN) and a Recurrent Neural Network (RNN) into one end-to-end trainable network. The former generates rich image representations, while the latter effectively analyzes long-term spatial dependencies. Besides, on top of the structure, we adopt an average pooling structure in order to deal with input images of arbitrary sizes. Experiments on several datasets, including SIW-13 and CVSI2015, demonstrate that our approach achieves superior performance, compared with previous approaches.

In this paper, we present a deep regression approach for face alignment. The deep regressor is a neural network that consists of a global layer and multistage local layers. The global layer estimates the initial face shape from the whole image, while the following local layers iteratively update the shape with local image observations. Combining standard derivations and numerical approximations, we make all layers able to backpropagate error differentials, so that we can apply the standard backpropagation to jointly learn the parameters from all layers. We show that the resulting deep regressor gradually and evenly approaches the true facial landmarks stage by stage, avoiding the tendency that often occurs in the cascaded regression methods and deteriorates the overall performance: yielding early stage regressors with high alignment accuracy gains but later stage regressors with low alignment accuracy gains. Experimental results on standard benchmarks demonstrate that our approach brings significant improvements over previous cascaded regression algorithms.

Aerial scene classification, which aims to automatically label an aerial image with a specific semantic category, is a fundamental problem for understanding high-resolution remote sensing imagery. In recent years, it has become an active task in the remote sensing area, and numerous algorithms have been proposed for this task, including many machine learning and data-driven approaches. However, the existing data sets for aerial scene classification, such as UC-Merced data set and WHU-RS19, contain relatively small sizes, and the results on them are already saturated. This largely limits the development of scene classification algorithms. This paper describes the Aerial Image data set (AID): a large-scale data set for aerial scene classification. The goal of AID is to advance the state of the arts in scene classification of remote sensing images. For creating AID, we collect and annotate more than 10000 aerial scene images. In addition, a comprehensive review of the existing aerial scene classification techniques as well as recent widely used deep learning methods is given. Finally, we provide a performance analysis of typical aerial scene classification and deep learning approaches on AID, which can be served as the baseline results on this benchmark.

Recognizing text in natural images is a challenging task with many unsolved problems. Different from those in documents, words in natural images often possess irregular shapes, which are caused by perspective distortion, curved character placement, etc. We propose RARE (Robust text recognizer with Automatic REctification), a recognition model that is robust to irregular text. RARE is a specially-designed deep neural network, which consists of a Spatial Transformer Network (STN) and a Sequence Recognition Network (SRN). In testing, an image is firstly rectified via a predicted Thin-Plate-Spline (TPS) transformation, into a more "readable" image for the following SRN, which recognizes text through a sequence recognition approach. We show that the model is able to recognize several types of irregular text, including perspective text and curved text. RARE is end-to-end trainable, requiring only images and associated text labels, making it convenient to train and deploy the model in practical systems. State-of-the-art or highly-competitive performance achieved on several benchmarks well demonstrates the effectiveness of the proposed model., Accepted by CVPR 2016

With the rapid growth of image and video data, there comes an interesting yet challenging problem: How to organize and utilize such large volume of data? Textual content in images and videos is an important source of information, which can be of great usefulness and assistance. Therefore, we investigate in this paper the problem of text image discrimination, which aims at distinguishing natural images with text from those without text. To tackle this problem, we propose a method that combines three mature techniques in this area, namely: MSER, CNN and BoW. To better evaluate the proposed algorithm, we also construct a large benchmark for text image discrimination, which includes natural images in a variety of scenarios. This algorithm has proven to be both effective and efficient, thus it can serve as a tool for mining valuable textual information from huge amount of image and video data.

With the rapid increase of transnational communication and cooperation, people frequently encounter multilingual scenarios in various situations. In this paper, we are concerned with a relatively new problem: script identification at word or line levels in natural scenes. A large-scale dataset with a great quantity of natural images and 10 types of widely used languages is constructed and released. In allusion to the challenges in script identification in real-world scenarios, a deep learning based algorithm is proposed. The experiments on the proposed dataset demonstrate that our algorithm achieves superior performance, compared with conventional image classification methods, such as the original CNN architecture and LLC., Comment: 5 pages, 7 figures, submitted to ICDAR 2015

Image-based sequence recognition has been a long-standing research topic in computer vision. In this paper, we investigate the problem of scene text recognition, which is among the most important and challenging tasks in image-based sequence recognition. A novel neural network architecture, which integrates feature extraction, sequence modeling and transcription into a unified framework, is proposed. Compared with previous systems for scene text recognition, the proposed architecture possesses four distinctive properties: (1) It is end-to-end trainable, in contrast to most of the existing algorithms whose components are separately trained and tuned. (2) It naturally handles sequences in arbitrary lengths, involving no character segmentation or horizontal scale normalization. (3) It is not confined to any predefined lexicon and achieves remarkable performances in both lexicon-free and lexicon-based scene text recognition tasks. (4) It generates an effective yet much smaller model, which is more practical for real-world application scenarios. The experiments on standard benchmarks, including the IIIT-5K, Street View Text and ICDAR datasets, demonstrate the superiority of the proposed algorithm over the prior arts. Moreover, the proposed algorithm performs well in the task of image-based music score recognition, which evidently verifies the generality of it., Comment: 5 figures

Driven by the wide range of applications, scene text detection and recognition have become active research topics in computer vision. Though extensively studied, localizing and reading text in uncontrolled environments remain extremely challenging, due to various interference factors. In this paper, we propose a novel multi-scale representation for scene text recognition. This representation consists of a set of detectable primitives, termed as strokelets, which capture the essential substructures of characters at different granularities. Strokelets possess four distinctive advantages: (1) Usability: automatically learned from bounding box labels, (2) Robustness: insensitive to interference factors, (3) Generality: applicable to variant languages, and (4) Expressivity: effective at describing characters. Extensive experiments on standard benchmarks verify the advantages of strokelets and demonstrate the effectiveness of the proposed algorithm for text recognition.