This inventor holds 2 USPTO granted patents and 1 published patent application. Top assignee: Google Inc.. Active years: 2026.
Company Filing History:
Years Active: 2026
Title: Suhani Deepak-Ranu Vora: Innovator in Geometry-Free Neural Scene Representations
Introduction
Suhani Deepak-Ranu Vora is an accomplished inventor based in San Mateo, CA. She has made significant contributions to the field of machine learning, particularly in the area of neural scene representations. Her innovative work has led to the development of a unique patent that enhances the way we synthesize images from different perspectives.
Latest Patents
Suhani holds a patent titled "Geometry-free neural scene representations through novel-view synthesis." This patent introduces machine learning models that generate geometry-free neural scene representations through efficient object-centric novel-view synthesis. The framework involves an encoder model, such as an encoder transformer network, which processes one or more RGB images to produce a fully latent scene representation. This representation can then be utilized by a decoder model to synthesize images from various target poses in a single forward pass. The use of transformers allows the encoder to learn an attention model that extracts essential 3D information from a limited set of images, enabling the rendering of novel views with accurate projections, parallax, occlusions, and semantics, all without explicit geometry.
Career Highlights
Suhani is currently employed at Google Inc., where she continues to push the boundaries of innovation in machine learning and computer vision. Her work is characterized by a strong focus on developing advanced algorithms that improve the efficiency and effectiveness of image synthesis.
Collaborations
Suhani has collaborated with notable colleagues, including Seyed Mohammad Mehdi Sajjadi and Noha Radwan. These collaborations have further enriched her research and contributed to the advancement of her projects.
Conclusion
Suhani Deepak-Ranu Vora is a pioneering inventor whose work in geometry-free neural scene representations is shaping the future of machine learning. Her innovative approaches and collaborations are paving the way for new advancements in the field.