Tag: deep learning

Abstract: In the computer vision literature, many effective histogram-based features have been developed. These engineered features include local binary patterns and edge histogram descriptors among others and they have been shown to be informative features for a variety of computer vision tasks. In this paper, we explore whether these features can be learned through histogram […]

Abstract: Emergent ecosystem properties, such as population and trait distributions, biodiversity and energy and water fluxes, occur because of the dynamic interactions of individuals in their environment. Remote sensing, where image data is collected over large areas, can provide information about individual organisms that reveals important ecosystem patterns and processes that are critical for macrosystems […]

Abstract: Triplet loss traditionally relies only on class labels and does not use all available information in multi-task scenarios where multiple types of annotations are available. This paper introduces a Multi-Annotation Triplet Loss (MATL) framework that extends triplet loss by incorporating additional annotations, such as bounding box information, alongside class labels in the loss formulation. […]

Congratulations to Dr. Matthew Cook for successfully passing his PhD dissertation exam! Dr. Cook’s research centers on developing algorithms that use null space projections within neural networks to add additional functionality. The primary application of this method is detecting out-of-distribution data, such as encountering unexpected objects during Automatic Target Recognition data collections. Moreover, results showed […]

Congratulations to Dr. Aditya Dutt for successfully passing his PhD dissertation exam! Dr. Dutt’s research introduced the Contrastive MultiModal Alignment Network (COMMANet), a novel approach to shared manifold-based domain translation and fusion. His work addressed the challenge of limited and imbalanced labeled datasets by leveraging contrastive learning with triplet networks to align multimodal data—such as […]

Congratulations to Dr. Spencer Chang for successfully passing his PhD dissertation exam! Dr. Chang’s research demonstrated the potential of combining learnable histogram features with deep learning convolutional methods. Additionally, he delivered key insights into enhancing statistical texture feature learning and explored the nuances of embedding histogram layers to maximize their impact within feature extraction pipelines. […]

Congratulations to our labmate Xiaolei Guo for successfully defending her dissertation! Defending a dissertation is the last milestone to completing a Ph.D. at the University of Florida. Xiaolei presented a deep interactive segmentation framework to address the time-consuming task of fine-scale pixel-level image annotation. Utilizing transfer learning, annotators are able to interactively fine-tune a pre-trained […]

It is a great pleasure and honor for everyone in Gatorsense that one of our labmates has achieved his goal. Congratulations to Dr. Connor McCurley for graduating with his Ph.D.! Connor’s dissertation is titled “Discriminative Feature Learning with Imprecise, Uncertain, and Ambiguous Data.” Connor developed approaches for discriminative representation learning using weakly-labeled ground-truth. He explored methods […]

Congratulations to our labmates and collaborators: Joshua Peeples, Sarah Walker, Connor McCurley, Alina Zare, James Keller and Weihuang Xu! Their paper, “Divergence Regulated Encoder Network for Joint Dimensionality Reduction and Classification”, was recently accepted to IEEE Geoscience and Remote Sensing Letters, 2022. In the paper, the authors investigate performing joint dimensionality reduction and classification using […]

Congratulations to our labmate Yiming Cui for successfully defending his research proposal!  Defending an oral research proposal is the second of four milestones to completing a Ph.D. at the University of Florida.  Yiming is planning to conduct point cloud semantic segmentation techniques using graph convolutional networks trained with weak annotations. We are excited to see […]