Research

Broadly, my research interests focus around new imaging technologies and computer vision algorithms that advance our understanding of the world around us: most importantly, CV/ML technologies that help both people and the planet.

Current Projects

Nutrition Logging - In collaboration with researchers at the Tufts School of Nutrition, I’ve been working on models that combine deep learning and classical computer vision techniques for estimating nutrient intake based on images of plated meals. I also supervised a team of senior undergraduate students completing a capstone project on semantic segmentation with direct applications to this work. Dataset and methods paper coming soon!

Hyperspectral Agriculture - I’ve been working on developing deep learning algorithms for hyperspectral imagery, specifically, band selection and dimensionality reduction techniques for agricultural robotics. I gave an oral presentation at SPIE DCS 2024 on classifying the phenological development stage of broccoli and cauliflower plants using these techniques.

Publications

Towards phenological development quantification in Brassica plants via multispectral imaging and deep learning
A. Stone, S. Rajeev, S.P. Rao, K. Panetta, S. Agaian
SPIE Multimodal Image Exploitation and Learning 2024 (SPIE DCS 2024)

Gaze depth estimation for eye-tracking systems
A. Stone, S. Rajeev, S.P. Rao, K. Panetta, S. Agaian, A. Gardony, J. Nordlund, R. Skantar
SPIE Multimodal Image Exploitation and Learning 2023 (SPIE DCS 2023)

A comprehensive 2D + 3D dataset for benchmarking hyperspectral imaging systems
A. Stone, S. P. Rao, S. Rajeev, K. Panetta, and S. Agaian
IEEE International Symposium on Technologies for Homeland Security (IEEE HST 2022, virtual presentation)

Gaze-FTNet: a feature transverse architecture for predicting gaze attention
S. Rajeev, S. K. KM, A. Stone, K. Panetta, and S. S. Agaian
SPIE Multimodal Image Exploitation and Learning 2022 (SPIE DCS 2022)