Computer Vision
Enabling computers and devices to understand what they see.
Publication
Simplifying DINO via Coding Rate Regularization
Podcast
Ideas: Building AI for population-scale systems with Akshay Nambi
Advances in AI are driving meaningful real-world impact. Principal Researcher Akshay Nambi shares how his passion for tackling real-world challenges across various domains fuels his work in building reliable and robust AI systems.
Publication
ε-VAE: Denoising as Visual Decoding
Microsoft Research Blog
Research Focus: Week of January 27, 2025
In this issue: A new approach to multimodal pretraining for remote sensing; Managed-retention memory for the AI era; Improving detection of macular telangiectasia type 2; Generalizing symbolic automata.
Dataset Source Code
RAS: Region-Adaptive Sampling for Diffusion Transformers
The RAS library is an open-source implementation of Regional-Adaptive Sampling (RAS), a novel diffusion model sampling strategy that introduces regional variability in sampling steps. Unlike conventional methods that uniformly process all image regions, RAS dynamically…