ICCV 2025
Novel view synthesis at scale
Training large vision transformers to create new views of a scene, using generated examples to expand the available training data.
View projectProjects
Selected work on large-scale training, generative models and multimodal learning.
18 peer-reviewed publications
Research internships at Google · NVIDIA · Adobe · MERL
Publications
9 research projects
CVPR 2025
GenDeg
A diffusion model creates degraded versions of clean images, giving restoration models more varied examples to learn from.
View project
CVPR 2023
Unite and Conquer
Generate images from combinations of text, class labels, semantic maps and sketches. Models learn each condition separately and combine them during sampling.
View project
ICCV 2023
Steered Diffusion
Guide a pretrained image generator to restore images or follow new conditions, without training a separate generator for each task.
View project
ECCV 2024
MaxFusion
Combine controls such as pose, depth and scene layout so they can guide the same image, even when parts of the controls disagree.
View project
Accepted to WACV 2027
Scale-Wise VAR
Show how a particular scale-by-scale image generator can behave like a diffusion model, allowing it to refine its predictions.
View project
FG 2025
Unconstrained face recognition
Train and adapt face-recognition models for images that are distant, small or degraded, where clean close-up training alone is not enough.
View project
WACV Oral 2025
PETALface
Adapt a strong face-recognition model to degraded images by training small quality-dependent adapters instead of the entire backbone.
View project
CVPR 2024
CrowdDiff
Generate several possible crowd maps and combine their head detections to reduce counting errors in dense or ambiguous scenes.
View projectInternships
Google · Dec 2024 – Mar 2025
Novel view synthesis at scale
Training large vision transformers to create new views of a scene, using generated examples to expand the available training data.
View project
NVIDIA · 2024
Physical-property assignment to 3D meshes
Use visual foundation models and multimodal reasoning to divide 3D objects into meaningful parts and estimate what those parts are made of.
View project
Adobe · May – Aug 2023
Control2Control
I worked on Control2Control during my Adobe internship. The project uses sketches, poses and layouts to edit an image while helping preserve the surrounding scene.
View projectMERL · May – Aug 2022
Steered Diffusion
Guide a pretrained image generator to restore images or follow new conditions, without training a separate generator for each task.
View projectSpecial thanks to Codex for helping create this Projects section.