Nonalcoholic fatty liver disease represents a major global burden of liver-related illnesses. Its progressive form, known as nonalcoholic steatohepatitis (Progressive NASH), is closely associated with moderate to advanced liver fibrosis. The FibroScan-AST score has been proposed as a non-invasive assessment tool to effectively identify high-risk patients who meet the criteria for progressive NASH. However, as the FAST score relies on blood test data, its applicability for real-time risk prediction remains limited.This study focuses on the high-precision segmentation of liver parenchyma in ultrasound images and proposes a modified U-Net architecture integrated with channel and spatial attention modules. The model is designed to accurately extract the liver parenchyma region while mitigating interference from rib shadows, intestinal gas, and background noise. To verify the practical applicability of the segmentation model, the FibroScan-AST classification task was incorporated. A weighted Dice loss was employed to address the class imbalance caused by the relatively small target area, and various data augmentation strategies were utilized to enhance generalization. The segmentation model achieved an average Dice score of 0.9 on the test set, demonstrating high accuracy in delineating liver parenchyma. Visual inspection further confirms that the predicted liver boundaries are clearer, with non-liver tissue effectively excluded, resulting in clean and semantically meaningful inputs for the downstream classification model.To validate the utility of the segmentation results, we further constructed a FAST classification model and compared the performance of three strategies: using raw images, using segmented ROIs, and an end-to-end integration of segmentation and classification. Experimental results show that using the segmented ROI as input significantly improves the F1-score (from 0.47 to 0.61), confirming that accurate liver segmentation can enhance downstream task performance and hold strong clinical application potential. This study verifies the necessity of high-quality liver parenchyma segmentation in ultrasound imaging and provides an image-driven framework for progressive NASH screening.

We introduce LoCoRe, Long-Context Re-ranker, a model that takes as input local descriptors corresponding to an image query and a list of gallery images and outputs similarity scores between the query and each gallery image. This model is used for image retrieval, where typically a first ranking is performed with an efficient similarity measure, and then a shortlist of top-ranked images is re-ranked based on a more fine-grained similarity model. Compared to existing methods that perform pair-wise similarity estimation with local descriptors or list-wise re-ranking with global descriptors, LoCoRe is the first method to perform list-wise re-ranking with local descriptors. To achieve this, we leverage efficient long-context sequence models to effectively capture the dependencies between query and gallery images at the local-descriptor level. During testing, we process long shortlists with a sliding window strategy that is tailored to overcome the context size limitations of sequence models. Our approach achieves superior performance compared with other re-rankers on established image retrieval benchmarks of landmarks ($\mathcal{R}{\text{Oxf}}$ and $\mathcal{R}{\text{Par}}$), products (SOP), fashion items (In-Shop), and bird species (CUB-200) while having comparable latency to the pair-wise local descriptor re-rankers.

With the advent of depth enabled sensors and increasing needs in surveillance systems, we propose a novel framework to detect fine-grained human attributes (e.g., having backpack, talking on cell phone, wearing glasses) in the surveillance environments. Traditional detection and recognition methods generally suffer from the problems such as variations in lighting conditions, poses, and viewpoints of object instances. To tackle these problems, we propose a multi-view part-based attribute detecting system based on color-depth inputs instead of purely utilizing color images. We address several important attributes in the surveillance environments and train multiple attribute classifiers based on features inferred from 3D information to construct our discriminative model. Several state-of-the-art methods are compared and the experimental results show that our method is more robust under large variations in surveillance conditions.

We propose a 3D sub-query expansion approach for boosting sketch-based multi-view image retrieval. The core idea of our method is to automatically convert two (guided) 2D sketches into an approximated 3D sketch model, and then generate multi-view sketches as expanded sub-queries to improve the retrieval performance. To learn the weights among synthesized views (sub-queries), we present a new multi-query feature to model the similarity between sub-queries and dataset images, and formulate it into a convex optimization problem. Our approach shows superior performance compared with the state-of-the-art approach on a public multi-view image dataset. Moreover, we also conduct sensitivity tests to analyze the parameters of our approach based on the gathered user sketches.