Exploring spatial contextual information is a well-adopted approach to achieving better semantic segmentation performance. However, most existing methods neglect the class association between the neighboring pixels. In this paper, we propose a CACINet, which consists of a Semantic Affinity Module (SAM) and a Class Association Module (CAM), to generate class-aware contextual information among pixels on a fine-grained level. SAM analyzes the affiliation of any two given pixels belonging to the same or different class. It produces intra-class and inter-class pixel contextual information. CAM classifies the image into different class regions globally and then it encodes the pixel based on the degree of affiliation of the pixels with each class in the image. In this way, it augments the class affiliation of the pixels into the corresponding context calculation. Comprehensive experiments demonstrate that the proposed method achieves competitive performance on two semantic segmentation benchmarks: ADE20K and PASCAL-Context.

Recent advances in vision transformers (ViTs) have achieved outstanding performance in visual recognition tasks, including image classification and detection. ViTs can learn global representations with their self-attention mechanism, but they are usually heavy-weight and unsuitable for resource-constrained devices. In this paper, we propose a novel linear feature attention (LFA) module to reduce computation costs for vision transformers and combine efficient mobile CNN modules to form a parameter-efficient and high-performance CNN-ViT hybrid model, called LightFormer, which can serve as a general-purpose backbone to learn both global and local representation. Comprehensive experiments demonstrate that LightFormer achieves competitive performance across different visual recognition tasks. On the ImageNet-1K dataset, LightFormer achieves top-1 accuracy of 78.5% with 5.5 million parameters. Our model also performs well when transferred to object detection and semantic segmentation tasks. On the MS COCO dataset, LightFormer attains mAP of 33.2 within the YOLOv3 framework, and on the Cityscapes dataset, with only a simple all-MLP decoder, LightFormer achieves mIoU of 78.5 and FPS of 15.3, surpassing state-of-the-art lightweight segmentation networks.

Deep neural networks executing with low precision at inference time can gain acceleration and compression advantages over their high-precision counterparts, but need to overcome the challenge of accuracy degeneration as the bit-width decreases. This work focuses on under 4-bit quantization that has a significant accuracy degeneration. We start with ternarization, a balance between efficiency and accuracy that quantizes both weights and activations into ternary values. We find that the hard threshold$\Delta $introduced in previous ternary networks for determining quantization intervals and the suboptimal solution of$\Delta $limit the performance of the ternary model. To alleviate it, we present Soft Threshold Ternary Networks (STTN), which enables the model to automatically determine ternarized values instead of depending on a hard threshold. Based on it, we further generalize the idea of soft threshold from ternarization to arbitrary bit-width, named Soft Threshold Quantized Networks (STQN). We observe that previous quantization relies on the rounding-to-nearest function, constraining the quantization solution space and leading to a significant accuracy degradation, especially in low-bit ($\leq3$-bits) quantization. Instead of relying on the traditional rounding-to-nearest function, STQN is able to determine quantization intervals by itself adaptively. Accuracy experiments on image classification, object detection and instance segmentation, as well as efficiency experiments on field-programmable gate array (FPGA) demonstrate that the proposed framework can achieve a prominent tradeoff between accuracy and efficiency. Code is available at:https://github.com/WeixiangXu/STTN.

Scene text detection attracts much attention in computer vision, because it can be widely used in many applications such as real-time text translation, automatic information entry, blind person assistance, robot sensing and so on. Though many methods have been proposed for horizontal and oriented texts, detecting irregular shape texts such as curved texts is still a challenging problem. To solve the problem, we propose a robust scene text detection method with adaptive text region representation. Given an input image, a text region proposal network is first used for extracting text proposals. Then, these proposals are verified and refined with a refinement network. Here, recurrent neural network based adaptive text region representation is proposed for text region refinement, where a pair of boundary points are predicted each time step until no new points are found. In this way, text regions of arbitrary shapes are detected and represented with adaptive number of boundary points. This gives more accurate description of text regions. Experimental results on five benchmarks, namely, CTW1500, TotalText, ICDAR2013, ICDAR2015 and MSRA-TD500, show that the proposed method achieves state-of-the-art in scene text detection.

Scene text detection is challenging as the input may have different orientations, sizes, font styles, lighting conditions, perspective distortions and languages. This paper addresses the problem by designing a Rotational Region CNN (R2CNN). R2CNN includes a Text Region Proposal Network (Text-RPN) to estimate approximate text regions and a multitask refinement network to get the precise inclined box. Our work has the following features. First, we use a novel multi-task regression method to support arbitrarily-oriented scene text detection. Second, we introduce multiple ROIPoolings to address the scene text detection problem for the first time. Third, we use an inclined Non-Maximum Suppression (NMS) to post-process the detection candidates. Experiments show that our method outperforms the state-of-the-art on standard benchmarks: ICDAR 2013, ICDAR 2015, COCO-Text and MSRA-TD500.

Text detection and recognition in scene images and videos attract much attention in computer vision recently. However, most existing text detection and recognition methods only focus on static images. In this paper an end-to-end scene text recognition method based on multi frame tracking is proposed for text in videos, in which temporal information is employed to improve performance. First, an end-to-end text recognition method based on a unified deep neural network is used to detect and recognize text in each frame of the input video. Then, multi frame text tracking is employed through associations of texts in current frame and several previous frames to obtain final results. Experiments on ICDAR datasets demonstrate that the proposed method outperforms the state-of-the-art methods in end-to-end video text recognition.

Scene text detection is a challenging problem in computer vision. In this paper, we propose a novel text detection network based on prevalent object detection frameworks. In order to obtain stronger semantic feature, we adopt ResNet as feature extraction layers and exploit multi-level feature by combining hierarchical convolutional networks. A vertical proposal mechanism is utilized to avoid proposal classification, while regression layer remains working to improve localization accuracy. Our approach evaluated on ICDAR2013 dataset achieves 0.91 F-measure, which outperforms previous state-of-the-art results in scene text detection.

Multinomial Naive Bayes with Expectation Maximization (MNB-EM) is a standard semi-supervised learning method to augment Multinomial Naive Bayes (MNB) for text classification. Despite its success, MNB-EM is not stable, and may succeed or fail to improve MNB. We believe that this is because MNB-EM lacks the ability to preserve the class distribution on words. In this paper, we propose a novel method to augment MNB-EM by leveraging the word-level statistical constraint to preserve the class distribution on words. The word-level statistical constraints are further converted to constraints on document posteriors generated by MNB-EM. Experiments demonstrate that our method can consistently improve MNB-EM, and outperforms state-of-art baselines remarkably.

Gesture inputs on multi-touch tabletops usually involve multiple fingers (more than two) and casual touchdowns or liftoffs of fingers. This flexibility of touch gestures allows more natural user interaction, but also poses new challenges for accurate recognition of multi-touch gestures. To address these challenges, we propose a new approach to recognize flexible multi-touch stroke gestures on tabletops. Based on a user study on multi-touch unistroke gestures, we develop a gesture recognition method by extracting key strokes embedded in flexible multi-touch input. Our evaluation study result shows that this method can greatly improve the recognition accuracy of flexible multi-touch unistroke gestures on tabletops.

With the fast economic growth and urbanization of many developing countries come concerns that their children now have fewer opportunities to express creativity and develop collaboration skills, or to experience their local cultural heritage. We propose to address these concerns by creating technologies inspired by traditional arts, and allowing children to create and collaborate through playing with them. ShadowStory is our first attempt in this direction, a digital storytelling system inspired by traditional Chinese shadow puppetry. We present the design and implementation of ShadowStory and a 7-day field trial in a primary school. Findings illustrated that ShadowStory promoted creativity, collaboration, and intimacy with traditional culture among children, as well as interleaved children's digital and physical playing experience.

Concept maps are an important tool to organize, represent, and share knowledge. Building a concept map involves creating text-based concepts and specifying their relationships with line-based links. Current concept map tools usually impose specific task structures for text and link construction, and may increase cognitive burden to generate and interact with concept maps. While pen-based devices (e.g., tablet PCs) offer users more freedom in drawing concept maps with a pen or stylus more naturally, the support for hand-drawn concept map creation and manipulation is still limited, largely due to the lack of methods to recognize the components and structures of hand-drawn concept maps. This article proposes a method to understand hand-drawn concept maps. Our algorithm can extract node blocks, or concept blocks, and link blocks of a hand-drawn concept map by combining dynamic programming and graph partitioning, recognize the text content of each concept node, and build a concept-map structure by relating concepts and links. We also design an algorithm for concept map retrieval based on hand-drawn queries. With our algorithms, we introduce structure-based intelligent manipulation techniques and ink-based retrieval techniques to support the management and modification of hand-drawn concept maps. Results from our evaluation study show high structure recognition accuracy in real time of our method, and good usability of intelligent manipulation and retrieval techniques.

Computer-based geometry systems have been widely used for teaching and learning, but largely based on mouse-and-keyboard interaction, these systems usually require users to draw figures by following strict task structures defined by menus, buttons, and mouse and keyboard actions. Pen-based designs offer a more natural way to develop geometry theorem proofs with hand-drawn figures and scripts. This paper describes a pen-based geometry theorem proving system that can effectively recognize hand-drawn figures and hand-written proof scripts, and accurately establish the correspondence between geometric components and proof steps. Our system provides dynamic and intelligent visual assistance to help users understand the process of proving and allows users to manipulate geometric components and proof scripts based on structures rather than strokes. The results from evaluation study show that our system is well perceived and users have high satisfaction with the accuracy of sketch recognition, the effectiveness of visual hints, and the efficiency of structure-based manipulation.

Concept maps are an important tool to knowledge organization, representation, and sharing. Most current concept map tools do not provide full support for hand-drawn concept map creation and manipulation, largely due to the lack of methods to recognize hand-drawn concept maps. This paper proposes a structure recognition method. Our algorithm can extract node blocks and link blocks of a hand-drawn concept map by combining dynamic programming and graph partitioning and then build a concept-map structure by relating extracted nodes and links. We also introduce structure-based intelligent manipulation technique of hand-drawn concept maps. Evaluation shows that our method has high structure recognition accuracy in real time, and the intelligent manipulation technique is efficient and effective.

Chinese text entry is challenging on mobile devices which rely on keypad input. Entering one character may require many key presses. This paper proposes a multimodal text entry technique for Chinese. In this method, Chinese user can enter Chinese text by simultaneously using the simplified phonemic input method named Jianpin with keypad and speech utterance. The key of the technique is a multimodal fusion algorithm, which synchronizes speech and keypad input and fuses redundant information from two modalities to get the best candidate. A preliminary evaluation shows that users appreciate this technique and it could reduce key presses and enhance the input efficiency.