In this paper we introduce variable resoultion 4-k meshes, a powerful structure for the representation of geometric objects at multiple levels of detail. It combines most properties of other related descriptions with ...
详细信息
the algorithm Radial Flow Model can generate a 3D closed triangular mesh which fits an unregistered range image. though complex meshes may be produced, they are restricted to be homeomorph to spheres. In this communic...
详细信息
In this work the architecture of an "animation machine" is proposed. Its main goals are the definition of the animation in a more abstract level and the simulation of the characters' behaviors to detect ...
详细信息
In this work, we extend a novel seed-based segmentation algorithm, which provides global optimum solutions according to a graph-cut measure, subject to high-level boundary constraints: the simultaneously handling of b...
详细信息
ISBN:
(纸本)9781509035687
In this work, we extend a novel seed-based segmentation algorithm, which provides global optimum solutions according to a graph-cut measure, subject to high-level boundary constraints: the simultaneously handling of boundary polarity and connectivity constraints. the proposed method incorporates the connectivity constraint in the Oriented image Foresting Transform (OIFT), ensuring the generation of connected objects, but such that the connection between its internal seeds is guaranteed to have a user-controllable minimum width. In other frameworks, such as the min-cut/max-flow algorithm, the connectivity constraint is known to lead to NP-hard problems. In contrast, our method conserves the low complexity of the OIFT algorithm. In the experiments, we show improved results for the segmentation of thin and elongated objects, for the same amount of user interaction. Our dataset of natural images with true segmentation is publicly available to the community.
We present a method to measure reflectance and texture of surfaces in a one step process. For later use in digital image synthesis it is mandatory to separate the gathered intensity values into these two parts to elim...
详细信息
Motivated by the inherent data scarcity in the medical domain, this work studies few-shot retinal disease classification, using the brazilian Multilabel Ophtalmological Dataset. We compare different network architectu...
详细信息
ISBN:
(纸本)9798350376043;9798350376036
Motivated by the inherent data scarcity in the medical domain, this work studies few-shot retinal disease classification, using the brazilian Multilabel Ophtalmological Dataset. We compare different network architectures and non-trivial data augmentations under the application of the Reptile Algorithm, conducting quantitative and qualitative analysis. Regarding the architectures, we observe that Swin outperforms ViT and ResNet. We also observe that clever data augmentations not only improve performance, but can also generate prediction confidence distributions that are more interpretable and trustworthy. Furthermore, pre-training the models with domain-specific data leads to superior ability of the models to detect the relevant patterns in the images. Code is available at ***/gabjp/few-shot-BRSET.
this work describes PhotoPixJ 2.0, a system for integration and execution of Digital imageprocessing algorithms. PhotoPixJ 2.0 is the migration of the Java 1.0 based PhotoPixJ to Java 2 platform. It provides a multi-...
详细信息
the image Foresting Transform (IFT) is a general framework to develop imageprocessing tools for a variety of tasks such as image segmentation, boundary tracking, morphological filtering, pixel clustering, among other...
详细信息
ISBN:
(纸本)9781538622193
the image Foresting Transform (IFT) is a general framework to develop imageprocessing tools for a variety of tasks such as image segmentation, boundary tracking, morphological filtering, pixel clustering, among others. the Differential image Foresting Transform (DIFT) comes in handy for scenarios where multiple iterations of IFT over the same image with small modifications on the input parameters are expected, reducing the processing complexity from linear to sublinear with respect to the number of pixels. In this paper, we propose an enhanced variant of the DIFT algorithm that avoids inconsistencies, when the connectivity function is not monotonically incremental. Our algorithm works withthe classical and non-classifical connectivity functions based on root position. Experiments were conducted on a superpixel task, showing a significant improvement to a state-of-the-art method.
this paper presents our research in imageprocessing (filters used for edges detection, color transformation, and distortion correction) using Artificial Neural Networks. We have developed a tool, Neuron Color, in ord...
详细信息
暂无评论