3 papers
cs.CV2026
Curriculum-Based Strategies for Efficient Cross-Domain Action Recognition
Emily Kim, Allen Wu, Jessica Hodgins
Despite significant progress in human action recognition, generalizing to diverse viewpoints remains a challenge. Most existing datasets are captured from ground-level perspectives…
cs.CV2025
TurboPortrait3D: Single-step diffusion-based fast portrait novel-view synthesis
Emily Kim, Julieta Martinez, Timur Bagautdinov +1
We introduce TurboPortrait3D: a method for low-latency novel-view synthesis of human portraits. Our approach builds on the observation that existing image-to-3D models for portrait…
cs.CV2024
Texture- and Shape-based Adversarial Attacks for Overhead Image Vehicle Detection
Mikael Yeghiazaryan, Sai Abhishek Siddhartha Namburu, Emily Kim +4
Detecting vehicles in aerial images is difficult due to complex backgrounds, small object sizes, shadows, and occlusions. Although recent deep learning advancements have improved o…