An official website of the United States government
Sketch-to-image is an important task to reduce the burden of creating a color image from scratch. Unlike previous sketch-to-image models, where the image is synthesized in an end-to-end manner, leading to an unnaturalistic image, we propose a method by decomposing the problem into subproblems to generate a more naturalistic and reasonable image. It first generates an intermediate output which is a semantic mask map from the input sketch through instance and semantic segmentation in two levels, background segmentation and foreground segmentation. Background segmentation is formed based on the context of the foreground objects. Then, the foreground segmentations are sequentially added to the created background segmentation. Finally, the generated mask map is fed into an image-to-image translation model to generate an image. Our proposed method works with 92 distinct classes. Compared to state-of-the-art sketch-to-image models, our proposed method outperforms the previous methods and generates better images.
more »
« less
S5: Sketch-to-image Synthesis via Scene and Size Sensing
Sketch-to-image synthesis method transforms a simple abstract black-and-white sketch into an image. Most sketch-to-image synthesis methods generate an image in an end-to-end manner, leading to generate a non-satisfactory result. The reason is that, in end-to-end models, the models generate images directly from the input sketches. Thus, with very abstract and complicated sketches, the models might struggle in generating naturalistic images due to the simultaneous focus on both factors: overall shape and fine-grained details. In this paper, we propose to divide the problem into subproblems. To this end, an intermediate output, which is a semantic mask map, is first generated from the input sketch via an instance and semantic segmentation. In the instance segmentation stage, the objects' sizes might be modified depending on the surrounding environment and their respective size prior to reflect reality and produce more realistic images. In the semantic seg-mentation stage, a background segmentation is first constructed based on the context of the detected objects. Various natural scenes are implemented for both indoor and outdoor scenes. Following this, a foreground segmentation process is commenced, where each detected object is semantically added into the constructed segmented background. Then, in the next stage, an image-to-image translation model is leveraged to convert the semantic mask map into a colored image. Finally, a post-processing stage is incorporated to further enhance the image result. Extensive experiments demonstrate the superiority of our proposed method over state-of-the-art methods.
more »
« less
- Award ID(s):
- 2025234
- PAR ID:
- 10521806
- Publisher / Repository:
- Institute of Electrical and Electronics Engineers
- Date Published:
- Journal Name:
- IEEE MultiMedia
- ISSN:
- 1070-986X
- Page Range / eLocation ID:
- 1 to 8
- Subject(s) / Keyword(s):
- Sketch-to-Image Image Generation Image Synthesis Scene and Size Sensing
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Training a semantic segmentation model requires large densely-annotated image datasets that are costly to obtain. Once the training is done, it is also difficult to add new object categories to such segmentation models. In this paper, we tackle the few-shot semantic segmentation problem, which aims to perform image segmentation task on unseen object categories merely based on one or a few support example(s). The key to solving this few-shot segmentation problem lies in effectively utilizing object information from support examples to separate target objects from the background in a query image. While existing methods typically generate object-level representations by averaging local features in support images, we demonstrate that such object representations are typically noisy and less distinguishing. To solve this problem, we design an object representation generator (ORG) module which can effectively aggregate local object features from support image( s) and produce better object-level representation. The ORG module can be embedded into the network and trained end-to-end in a weakly-supervised fashion without extra human annotation. We incorporate this design into a modified encoder-decoder network to present a powerful and efficient framework for few-shot semantic segmentation. Experimental results on the Pascal-VOC and MS-COCO datasets show that our approach achieves better performance compared to existing methods under both one-shot and five-shot settings.more » « less
-
Abstract Contemporary approaches to instance segmentation in cell science use 2D or 3D convolutional networks depending on the experiment and data structures. However, limitations in microscopy systems or efforts to prevent phototoxicity commonly require recording sub-optimally sampled data that greatly reduces the utility of such 3D data, especially in crowded sample space with significant axial overlap between objects. In such regimes, 2D segmentations are both more reliable for cell morphology and easier to annotate. In this work, we propose the projection enhancement network (PEN), a novel convolutional module which processes the sub-sampled 3D data and produces a 2D RGB semantic compression, and is trained in conjunction with an instance segmentation network of choice to produce 2D segmentations. Our approach combines augmentation to increase cell density using a low-density cell image dataset to train PEN, and curated datasets to evaluate PEN. We show that with PEN, the learned semantic representation in CellPose encodes depth and greatly improves segmentation performance in comparison to maximum intensity projection images as input, but does not similarly aid segmentation in region-based networks like Mask-RCNN. Finally, we dissect the segmentation strength against cell density of PEN with CellPose on disseminated cells from side-by-side spheroids. We present PEN as a data-driven solution to form compressed representations of 3D data that improve 2D segmentations from instance segmentation networks.more » « less
-
Image synthesis is a process of converting the input text, sketch, or other sources, i.e., another image or mask, into an image. It is an important problem in the computer vision field, where it has attracted the research community to attempt to solve this challenge at a high level to generate photorealistic images. Different techniques and strategies have been employed to achieve this purpose. Thus, the aim of this paper is to provide a comprehensive review of various image synthesis models covering several aspects. First, the image synthesis concept is introduced. We then review different image synthesis methods divided into three categories: image generation from text, sketch, and other inputs, respectively. Each sub-category is introduced under the proper category based upon the general framework to provide a broad vision of all existing image synthesis methods. Next, brief details of the benchmarked datasets used in image synthesis are discussed along with specifying the image synthesis models that leverage them. Regarding the evaluation, we summarize the metrics used to evaluate the image synthesis models. Moreover, a detailed analysis based on the evaluation metrics of the results of the introduced image synthesis is provided. Finally, we discuss some existing challenges and suggest possible future research directions.more » « less
-
A self-driving car must be able to reliably handle adverse weather conditions (e.g., snowy) to operate safely. In this paper, we investigate the idea of turning sensor inputs (i.e., images) captured in an adverse condition into a benign one (i.e., sunny), upon which the downstream tasks (e.g., semantic segmentation) can attain high accuracy. Prior work primarily formulates this as an unpaired image-to-image translation problem due to the lack of paired images captured under the exact same camera poses and semantic layouts. While perfectly- aligned images are not available, one can easily obtain coarsely- paired images. For instance, many people drive the same routes daily in both good and adverse weather; thus, images captured at close-by GPS locations can form a pair. Though data from repeated traversals are unlikely to capture the same foreground objects, we posit that they provide rich contextual information to supervise the image translation model. To this end, we propose a novel training objective leveraging coarsely- aligned image pairs. We show that our coarsely-aligned training scheme leads to a better image translation quality and improved downstream tasks, such as semantic segmentation, monocular depth estimation, and visual localization.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.1109/MMUL.2024.3375610
