Qiyang Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qiyang Hu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Matthias Zwicker
University of Maryland College Park
6 shared publications - Tiziano Portenier
University of Bern Switzerland
4 shared publications - Paolo Favaro
University of Bern, Bern, Switzerland
3 shared publications - Paolo Favaro
University of Bern Switzerland
3 shared publications - Attila Szabó
University of Bern
2 shared publications - Attila Szabó
University of Bern Switzerland
2 shared publications
Latest Publications
- Open Access
Smart, Deep Copy-Paste
In this work, we propose a novel system for smart copy-paste, enabling the synthesis of high-quality results given a masked source image content and a target image context as input. Our system naturally resolves both shading and geometric inconsistencies between source and target image, resulting in a merged result image that features the content from the pasted source image, seamlessly pasted into the target context. Our framework is based on a novel training image transformation procedure that allows to train a deep convolutional neural network end-to-end to automatically learn a representation that is suitable for copy-pasting. Our training procedure works with any image dataset without additional information such as labels, and we demonstrate the effectiveness of our system on two popular datasets, high-resolution face images and the more complex Cityscapes dataset. Our technique outperforms the current state of the art on face images, and we show promising results on the Cityscapes dataset, demonstrating that our system generalizes to much higher resolution than the training data.
Open AccessSmart, Deep Copy-Paste
In this work, we propose a novel system for smart copy-paste, enabling the synthesis of high-quality results given a masked source image content and a target image context as input. Our system naturally resolves both shading and geometric inconsistencies between source and target image, resulting in a merged result image that features the content from the pasted source image, seamlessly pasted into the target context. Our framework is based on a novel training image transformation procedure that allows to train a deep convolutional neural network end-to-end to automatically learn a representation that is suitable for copy-pasting. Our training procedure works with any image dataset without additional information such as labels, and we demonstrate the effectiveness of our system on two popular datasets, high-resolution face images and the more complex Cityscapes dataset. Our technique outperforms the current state of the art on face images, and we show promising results on the Cityscapes dataset, demonstrating that our system generalizes to much higher resolution than the training data.
- 6 October 2018
- Published by Springer Nature
- p. 721-736
Understanding Degeneracies and Ambiguities in Attribute Transfer
- 6 October 2018
- Published by Springer Nature
- p. 721-736
Understanding Degeneracies and Ambiguities in Attribute Transfer
- Open Access
- 30 July 2018
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 37 (4) , 1-13
Faceshop
We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.
- 30 July 2018
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 37 (4) , 1-13
Open AccessFaceshop
We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.
- Open Access
FaceShop: Deep Sketch-based Face Image Editing
We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.
Open AccessFaceShop: Deep Sketch-based Face Image Editing
We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.
- Open Access
Challenges in Disentangling Independent Factors of Variation
We study the problem of building models that disentangle independent factors of variation. Such models could be used to encode features that can efficiently be used for classification and to transfer attributes between different images in image synthesis. As data we use a weakly labeled training set. Our weak labels indicate what single factor has changed between two data samples, although the relative value of the change is unknown. This labeling is of particular interest as it may be readily available without annotation costs. To make use of weak labels we introduce an autoencoder model and train it through constraints on image pairs and triplets. We formally prove that without additional knowledge there is no guarantee that two images with the same factor of variation will be mapped to the same feature. We call this issue the reference ambiguity. Moreover, we show the role of the feature dimensionality and adversarial training. We demonstrate experimentally that the proposed model can successfully transfer attributes on several datasets, but show also cases when the reference ambiguity occurs.
Open AccessChallenges in Disentangling Independent Factors of Variation
We study the problem of building models that disentangle independent factors of variation. Such models could be used to encode features that can efficiently be used for classification and to transfer attributes between different images in image synthesis. As data we use a weakly labeled training set. Our weak labels indicate what single factor has changed between two data samples, although the relative value of the change is unknown. This labeling is of particular interest as it may be readily available without annotation costs. To make use of weak labels we introduce an autoencoder model and train it through constraints on image pairs and triplets. We formally prove that without additional knowledge there is no guarantee that two images with the same factor of variation will be mapped to the same feature. We call this issue the reference ambiguity. Moreover, we show the role of the feature dimensionality and adversarial training. We demonstrate experimentally that the proposed model can successfully transfer attributes on several datasets, but show also cases when the reference ambiguity occurs.
- Open Access
- 29 July 2017
- Published by Association for Computing Machinery (ACM)
- p. 1-12
SmartSketcher
We present a sketch-based image retrieval system, designed to answer arbitrary queries that may go beyond searching for predefined object or scene categories. While sketching is fast and intuitive to formulate visual queries, pure sketch-based image retrieval often returns many outliers because it lacks a semantic understanding of the query. Our key idea is to combine sketch-based queries with inter-active, semantic re-ranking of query results. We leverage progress in deep learning and use a feature representation learned for image classification for re-ranking. This allows us to cluster semantically similar images, re-rank based on the clusters, and present more meaningful query results to the user. We report on two large-scale benchmarks and demonstrate that our re-ranking approach leads to significant improvements over the state of the art. Finally, a user study designed to evaluate a practical use case confirms the benefits of our approach.
- 29 July 2017
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessSmartSketcher
We present a sketch-based image retrieval system, designed to answer arbitrary queries that may go beyond searching for predefined object or scene categories. While sketching is fast and intuitive to formulate visual queries, pure sketch-based image retrieval often returns many outliers because it lacks a semantic understanding of the query. Our key idea is to combine sketch-based queries with inter-active, semantic re-ranking of query results. We leverage progress in deep learning and use a feature representation learned for image classification for re-ranking. This allows us to cluster semantically similar images, re-rank based on the clusters, and present more meaningful query results to the user. We report on two large-scale benchmarks and demonstrate that our re-ranking approach leads to significant improvements over the state of the art. Finally, a user study designed to evaluate a practical use case confirms the benefits of our approach.