Intrinsic image decomposition (IID) is an under-constrained problem. Therefore, traditional approaches use hand crafted priors to constrain the problem. However, these constraints are limited when coping with complex scenes. Deep learning-based approaches learn these constraints implicitly through the data, but they often suffer from dataset biases (due to not being able to include all possible imaging conditions). In this paper, a combination of the two is proposed. Component specific priors like semantics and invariant features are exploited to obtain semantically and physically plausible reflectance transitions. These transitions are used to steer a progressive CNN with implicit homogeneity constraints to decompose reflectance and shading maps. An ablation study is conducted showing that the use of the proposed priors and progressive CNN increase the IID performance. State of the art performance on both our proposed dataset and the standard real-world IIW dataset shows the effectiveness of the proposed method. Code is made available at https://github.com/Morpheus3000/SIGNet
Artificial intelligence is currently powering diverse real-world applications. These applications have shown promising performance, but raise complicated ethical issues, i.e. how to embed ethics to make AI applications behave morally. One way toward moral AI systems is by imitating human prosocial behavior and encouraging some form of good behavior in systems. However, learning such normative ethics (especially from images) is challenging mainly due to a lack of data and labeling complexity. Here, we propose a model that predicts visual commonsense immorality in a zero-shot manner. We train our model with an ETHICS dataset (a pair of text and morality annotation) via a CLIP-based image-text joint embedding. In a testing phase, the immorality of an unseen image is predicted. We evaluate our model with existing moral/immoral image datasets and show fair prediction performance consistent with human intuitions. Further, we create a visual commonsense immorality benchmark with more general and extensive immoral visual contents. Codes and dataset are available at https://github.com/ku-vai/Zero-shot-Visual-Commonsense-Immorality-Prediction. Note that this paper might contain images and descriptions that are offensive in nature.
World-wide-web, with the website and webpage as the main interface, facilitates the dissemination of important information. Hence it is crucial to optimize them for better user interaction, which is primarily done by analyzing users' behavior, especially users' eye-gaze locations. However, gathering these data is still considered to be labor and time intensive. In this work, we enable the development of automatic eye-gaze estimations given a website screenshots as the input. This is done by the curation of a unified dataset that consists of website screenshots, eye-gaze heatmap and website's layout information in the form of image and text masks. Our pre-processed dataset allows us to propose an effective deep learning-based model that leverages both image and text spatial location, which is combined through attention mechanism for effective eye-gaze prediction. In our experiment, we show the benefit of careful fine-tuning using our unified dataset to improve the accuracy of eye-gaze predictions. We further observe the capability of our model to focus on the targeted areas (images and text) to achieve high accuracy. Finally, the comparison with other alternatives shows the state-of-the-art result of our model establishing the benchmark for the eye-gaze prediction task.
During image editing, existing deep generative models tend to re-synthesize the entire output from scratch, including the unedited regions. This leads to a significant waste of computation, especially for minor editing operations. In this work, we present Spatially Sparse Inference (SSI), a general-purpose technique that selectively performs computation for edited regions and accelerates various generative models, including both conditional GANs and diffusion models. Our key observation is that users tend to make gradual changes to the input image. This motivates us to cache and reuse the feature maps of the original image. Given an edited image, we sparsely apply the convolutional filters to the edited regions while reusing the cached features for the unedited regions. Based on our algorithm, we further propose Sparse Incremental Generative Engine (SIGE) to convert the computation reduction to latency reduction on off-the-shelf hardware. With 1.2%-area edited regions, our method reduces the computation of DDIM by 7.5$\times$ and GauGAN by 18$\times$ while preserving the visual fidelity. With SIGE, we accelerate the speed of DDIM by 3.0x on RTX 3090 and 6.6$\times$ on Apple M1 Pro CPU, and GauGAN by 4.2$\times$ on RTX 3090 and 14$\times$ on Apple M1 Pro CPU.
Dunhuang murals are a collection of Chinese style and national style, forming a self-contained Chinese-style Buddhist art. It has very high historical and cultural value and research significance. Among them, the lines of Dunhuang murals are highly general and expressive. It reflects the character's distinctive character and complex inner emotions. Therefore, the outline drawing of murals is of great significance to the research of Dunhuang Culture. The contour generation of Dunhuang murals belongs to image edge detection, which is an important branch of computer vision, aims to extract salient contour information in images. Although convolution-based deep learning networks have achieved good results in image edge extraction by exploring the contextual and semantic features of images. However, with the enlargement of the receptive field, some local detail information is lost. This makes it impossible for them to generate reasonable outline drawings of murals. In this paper, we propose a novel edge detector based on self-attention combined with convolution to generate line drawings of Dunhuang murals. Compared with existing edge detection methods, firstly, a new residual self-attention and convolution mixed module (Ramix) is proposed to fuse local and global features in feature maps. Secondly, a novel densely connected backbone extraction network is designed to efficiently propagate rich edge feature information from shallow layers into deep layers. Compared with existing methods, it is shown on different public datasets that our method is able to generate sharper and richer edge maps. In addition, testing on the Dunhuang mural dataset shows that our method can achieve very competitive performance.
In order to be able to use artificial intelligence (AI) in medicine without scepticism and to recognise and assess its growing potential, a basic understanding of this topic is necessary among current and future medical staff. Under the premise of "trust through understanding", we developed an innovative online course as a learning opportunity within the framework of the German KI Campus (AI campus) project, which is a self-guided course that teaches the basics of AI for the analysis of medical image data. The main goal is to provide a learning environment for a sufficient understanding of AI in medical image analysis so that further interest in this topic is stimulated and inhibitions towards its use can be overcome by means of positive application experience. The focus was on medical applications and the fundamentals of machine learning. The online course was divided into consecutive lessons, which include theory in the form of explanatory videos, practical exercises in the form of Streamlit and practical exercises and/or quizzes to check learning progress. A survey among the participating medical students in the first run of the course was used to analyse our research hypotheses quantitatively.
Arbitrary style transfer (AST) transfers arbitrary artistic styles onto content images. Despite the recent rapid progress, existing AST methods are either incapable or too slow to run at ultra-resolutions (e.g., 4K) with limited resources, which heavily hinders their further applications. In this paper, we tackle this dilemma by learning a straightforward and lightweight model, dubbed MicroAST. The key insight is to completely abandon the use of cumbersome pre-trained Deep Convolutional Neural Networks (e.g., VGG) at inference. Instead, we design two micro encoders (content and style encoders) and one micro decoder for style transfer. The content encoder aims at extracting the main structure of the content image. The style encoder, coupled with a modulator, encodes the style image into learnable dual-modulation signals that modulate both intermediate features and convolutional filters of the decoder, thus injecting more sophisticated and flexible style signals to guide the stylizations. In addition, to boost the ability of the style encoder to extract more distinct and representative style signals, we also introduce a new style signal contrastive loss in our model. Compared to the state of the art, our MicroAST not only produces visually superior results but also is 5-73 times smaller and 6-18 times faster, for the first time enabling super-fast (about 0.5 seconds) AST at 4K ultra-resolutions. Code is available at https://github.com/EndyWon/MicroAST.
Contrastive models like CLIP have been shown to learn robust representations of images that capture both semantics and style. To leverage these representations for image generation, we propose a two-stage model: a prior that generates a CLIP image embedding given a text caption, and a decoder that generates an image conditioned on the image embedding. We show that explicitly generating image representations improves image diversity with minimal loss in photorealism and caption similarity. Our decoders conditioned on image representations can also produce variations of an image that preserve both its semantics and style, while varying the non-essential details absent from the image representation. Moreover, the joint embedding space of CLIP enables language-guided image manipulations in a zero-shot fashion. We use diffusion models for the decoder and experiment with both autoregressive and diffusion models for the prior, finding that the latter are computationally more efficient and produce higher-quality samples.
Out-Of-Distribution (OOD) detection has received broad attention over the years, aiming to ensure the reliability and safety of deep neural networks (DNNs) in real-world scenarios by rejecting incorrect predictions. However, we notice a discrepancy between the conventional evaluation vs. the essential purpose of OOD detection. On the one hand, the conventional evaluation exclusively considers risks caused by label-space distribution shifts while ignoring the risks from input-space distribution shifts. On the other hand, the conventional evaluation reward detection methods for not rejecting the misclassified image in the validation dataset. However, the misclassified image can also cause risks and should be rejected. We appeal to rethink OOD detection from a human-centric perspective, that a proper detection method should reject the case that the deep model's prediction mismatches the human expectations and adopt the case that the deep model's prediction meets the human expectations. We propose a human-centric evaluation and conduct extensive experiments on 45 classifiers and 8 test datasets. We find that the simple baseline OOD detection method can achieve comparable and even better performance than the recently proposed methods, which means that the development in OOD detection in the past years may be overestimated. Additionally, our experiments demonstrate that model selection is non-trivial for OOD detection and should be considered as an integral of the proposed method, which differs from the claim in existing works that proposed methods are universal across different models.
People say, "A picture is worth a thousand words". Then how can we get the rich information out of the image? We argue that by using visual clues to bridge large pretrained vision foundation models and language models, we can do so without any extra cross-modal training. Thanks to the strong zero-shot capability of foundation models, we start by constructing a rich semantic representation of the image (e.g., image tags, object attributes / locations, captions) as a structured textual prompt, called visual clues, using a vision foundation model. Based on visual clues, we use large language model to produce a series of comprehensive descriptions for the visual content, which is then verified by the vision model again to select the candidate that aligns best with the image. We evaluate the quality of generated descriptions by quantitative and qualitative measurement. The results demonstrate the effectiveness of such a structured semantic representation.