domain adaptation

Wetland ecosystems are affected by climate change, tidal fluctuations, and human activities, causing significant spectral variations across different times and locations. These variations challenge hyperspectral classification models, limiting their generalization in cross-domain scenarios. However, most existing hyperspectral datasets focus on a single time point or region, lacking standardized resources for cross-temporal and cross-scene studies, which restricts the application of unsupervised domain adaptation (UDA) methods in wetland remote sensing.

We evaluate our approach on three popular domain adaptation benchmark datasets. The first one is Office-Caltech10 dataset, which contains images of 10 object categories from an office environment (e.g., keyboard, laptop) in 4 sources: Amazon, Caltech256, DSLR, and Webcam. We encode each source into 4096-dimensional feature vectors. Using each source as a domain, we get four domains leading to 12 domain adaptation tasks. The second one is Office-Home dataset, which contains images of 65 object categories found typically in Office and Home settings.

This project investigates bias in automatic facial recognition (FR). Specifically, subjects are grouped into predefined subgroups based on gender, ethnicity, and age. We propose a novel image collection called Balanced Faces in the Wild (BFW), which is balanced across eight subgroups (i.e., 800 face images of 100 subjects, each with 25 face samples).

In this paper, we propose a framework for 3D human pose estimation using a single 360° camera mounted on the user's wrist. Perceiving a 3D human pose with such a simple setup has remarkable potential for various applications (e.g., daily-living activity monitoring, motion analysis for sports training). However, no existing method has tackled this task due to the difficulty of estimating a human pose from a single camera image in which only a part of the human body is captured, and because of a lack of training data.