Inappropriate visual content on the internet has spread everywhere, and thus children are exposed unintentionally to sexually explicit visual content. Animated cartoon movies sometimes have sensitive content such as pornography and sex. Usually, video sharing platforms take children’s e-safety into consideration through manual censorship, which is both time-consuming and expensive. Therefore, automated cartoon censorship is highly recommended to be integrated into media platforms. In this paper, various methods and approaches were explored to detect inappropriate visual content in cartoon animation. First, state-of-the-art conventional feature techniques were utilised and evaluated. In addition, a simple end-to-end convolutional neural network (CNN) was used and was found to outperform conventional techniques in terms of accuracy (85.33%) and F1 score (83.46%). Additionally, to target the deeper version of CNNs, ResNet, and EfficientNet were demonstrated and compared. The CNN-based extracted features were mapped into two classes: normal and porn. To improve the model’s performance, we utilised feature and decision fusion approaches which were found to outperform state-of-the-art techniques in terms of accuracy (87.87%), F1 score (87.87%), and AUC (94.40%). To validate the domain generalisation performance of the proposed methods, CNNs, pre-trained on the cartoon dataset were evaluated on public NPDI-800 natural videos and found to provide an accuracy of 79.92%, and F1 score of 80.58%. Similarly, CNNs, pre-trained on the public NPDI-800 natural videos, were evaluated on cartoon dataset and found to give an accuracy of 82.666%, and F1 score of 81.588%. Finally, a novel cartoon pornography dataset, with various characters, skin colours, positions, viewpoints, and scales, was proposed.

/pdf/an-evaluation-of-traditional-and-cnn-based-feature-3rgflwo7c8.pdf

An Evaluation of Traditional and CNN-Based Feature Descriptors for Cartoon Pornography Detection

Cartoon face recognition is challenging as they typically have smooth color regions and emphasized edges, the key to recognize cartoon faces is to precisely perceive their sparse and critical shape patterns. However, it is quite difficult to learn a shape-oriented representation for cartoon face recognition with convolutional neural networks (CNNs). To mitigate this issue, we propose the GraphJigsaw that constructs jigsaw puzzles at various stages in the classification network and solves the puzzles with the graph convolutional network (GCN) in a progressive manner. Solving the puzzles requires the model to spot the shape patterns of the cartoon faces as the texture information is quite limited. The key idea of GraphJigsaw is constructing a jigsaw puzzle by randomly shuffling the intermediate convolutional feature maps in the spatial dimension and exploiting the GCN to reason and recover the correct layout of the jigsaw fragments in a self-supervised manner. The proposed GraphJigsaw avoids training the classification model with the deconstructed images that would introduce noisy patterns and are harmful for the final classification. Specially, GraphJigsaw can be incorporated at various stages in a top-down manner within the classification model, which facilitates propagating the learned shape patterns gradually. GraphJigsaw does not rely on any extra manual annotation during the training process and incorporates no extra computation burden at inference time. Both quantitative and qualitative experimental results have verified the feasibility of our proposed GraphJigsaw, which consistently outperforms other face recognition or jigsaw-based methods on two popular cartoon face datasets with considerable improvements.

Graph Jigsaw Learning for Cartoon Face Recognition.

Cartoons and animation domain videos have very different characteristics compared to real‐life images and videos. In addition, this domain carries a large variability in styles. Current computer vision and deep‐learning solutions often fail on animated content because they were trained on natural images. In this paper we present a method to refine a semantic representation suitable for specific animated content. We first train a neural network on a large‐scale set of animation videos and use the mapping to deep features as an embedding space. Next, we use self‐supervision to refine the representation for any specific animation style by gathering many examples of animated characters in this style, using a multi‐object tracking. These examples are used to define triplets for contrastive loss training. The refined semantic space allows better clustering of animated characters even when they have diverse manifestations. Using this space we can build dictionaries of characters in an animation videos, and define specialized classifiers for specific stylistic content (e.g., characters in a specific animation series) with very little user effort. These classifiers are the basis for automatically labeling characters in animation videos. We present results on a collection of characters in a variety of animation styles.

CAST: Character labeling in Animation using Self‐supervision by Tracking

Cartoon face recognition is challenging as they typically have smooth color regions and emphasized edges, the key to recognize cartoon faces is to precisely perceive their sparse and critical shape patterns. However, it is quite difficult to learn a shape-oriented representation for cartoon face recognition with convolutional neural networks (CNNs). To mitigate this issue, we propose the GraphJigsaw that constructs jigsaw puzzles at various stages in the classification network and solves the puzzles with the graph convolutional network (GCN) in a progressive manner. Solving the puzzles requires the model to spot the shape patterns of the cartoon faces as the texture information is quite limited. The key idea of GraphJigsaw is constructing a jigsaw puzzle by randomly shuffling the intermediate convolutional feature maps in the spatial dimension and exploiting the GCN to reason and recover the correct layout of the jigsaw fragments in a self-supervised manner. The proposed GraphJigsaw avoids training the classification model with the deconstructed images that would introduce noisy patterns and are harmful for the final classification. Specially, GraphJigsaw can be incorporated at various stages in a top-down manner within the classification model, which facilitates propagating the learned shape patterns gradually. GraphJigsaw does not rely on any extra manual annotation during the training process and incorporates no extra computation burden at inference time. Both quantitative and qualitative experimental results have verified the feasibility of our proposed GraphJigsaw, which consistently outperforms other face recognition or jigsaw-based methods on two popular cartoon face datasets with considerable improvements. 

/pdf/graph-jigsaw-learning-for-cartoon-face-recognition-15xrvez6.pdf

Graph Jigsaw Learning for Cartoon Face Recognition

Humans have long been recorded in a variety of forms since antiquity. For example, sculptures and paintings were the primary media for depicting human beings before the invention of cameras. However, most current human-centric computer vision tasks like human pose estimation and human image generation focus exclusively on natural images in the real world. Artificial humans, such as those in sculptures, paintings, and cartoons, are commonly neglected, making existing models fail in these scenarios. As an abstraction of life, art incorporates humans in both natural and artificial scenes. We take advantage of it and introduce the Human-Art dataset to bridge related tasks in natural and artificial scenarios. Specifically, Human-Art contains 50k high-quality images with over 123k person instances from 5 natural and 15 artificial scenarios, which are annotated with bounding boxes, keypoints, self-contact points, and text information for humans represented in both 2D and 3D. It is, therefore, comprehensive and versatile for various downstream tasks. We also provide a rich set of baseline results and detailed analyses for related tasks, including human detection, 2D and 3D human pose estimation, image generation, and motion transfer. As a challenging dataset, we hope Human-Art can provide insights for relevant research and open up new research questions.

Human-Art: A Versatile Human-Centric Dataset Bridging Natural and Artificial Scenes

Recent years have witnessed increasing attention in cartoon media, powered by the strong demands of industrial applications. As the first step to understand this media, cartoon face recognition is a crucial but less-explored task with few datasets proposed. In this work, we first present a new challenging benchmark dataset, consisting of 389,678 images of 5,013 cartoon characters annotated with identity, bounding box, pose, and other auxiliary attributes. The dataset, named iCartoonFace, is currently the largest-scale, high-quality, rich-annotated, and spanning multiple occurrences in the field of image recognition, including near-duplications, occlusions, and appearance changes. In addition, we provide two types of annotations for cartoon media, i.e., face recognition, and face detection, with the help of a semi-automatic labeling algorithm. To further investigate this challenging dataset, we propose a multi-task domain adaptation approach that jointly utilizes the human and cartoon domain knowledge with three discriminative regularizations. We hence perform a benchmark analysis of the proposed dataset and verify the superiority of the proposed approach in the cartoon face recognition task. The dataset is available at https://iqiyi.cn/icartoonface.

Cartoon Face Recognition: A Benchmark Dataset

Recent years have witnessed increasing attention in cartoon media, powered by the strong demands of industrial applications. As the first step to understand this media, cartoon face recognition is a crucial but less-explored task with few datasets proposed. In this work, we first present a new challenging benchmark dataset, consisting of 389,678 images of 5,013 cartoon characters annotated with identity, bounding box, pose, and other auxiliary attributes. The dataset, named iCartoonFace, is currently the largest-scale, high-quality, richannotated, and spanning multiple occurrences in the field of image recognition, including near-duplications, occlusions, and appearance changes. In addition, we provide two types of annotations for cartoon media, i.e., face recognition, and face detection, with the help of a semi-automatic labeling algorithm. To further investigate this challenging dataset, we propose a multi-task domain adaptation approach that jointly utilizes the human and cartoon domain knowledge with three discriminative regularizations. We hence perform a benchmark analysis of the proposed dataset and verify the superiority of the proposed approach in the cartoon face recognition task. We believe this public availability will attract more research attention in broad practical application scenarios.

The rapid development of the logistics industry has brought a lot of conveniences. However, the stevedoring and transportation links are still manual working, with high labor intensity, low transportation efficiency, high cost, and prone to violence, resulting in cargo damage and other defects. This paper designs an intelligent stevedoring transport system based on visual recognition (ISTS-VR), which can adsorb express parcels of different sizes. Through four adjustable distributed suction cups, the force of express parcels is balanced in the lifting process to avoid sliding and dumping accidents. It can replace the clamp manipulator to prevent the packaging damage of items. According to the established 3D model and static simulation analysis, the system has theoretical operability.

Mengyuan Ren

Papers

Cartoon Face Recognition: A Benchmark Dataset

Cartoon Face Recognition: A Benchmark Dataset

Intelligent stevedoring transport system based on visual recognition