Research
We publish what we learn.
We test our capture and annotation methods on public foundation and robot models and publish the results. Below are our own papers, the granted patents behind the Alia optics, and the earlier computer vision work our team brought to the company.
66.6%
error reduction
Cosmos-Reason2-2B · PRISM
2.19×
improvement
GR00T N1.6 · SABER
6 / 7
metrics won
Video world models · RetailSMV
Measured on public foundation and robot models. Published, with reproducible results.
Publications
04Papers from the DreamVu research team.
2026
RetailSMV: Exocentric vs. Egocentric Adaptation of Foundation Video World Models in Retail
A. Rouhi, R. Aggarwal, P. Sakurikar, A. M. Namboodiri, S. P. Reddi
2026
PRISM-World: Teaching Video World Models the Physics of Retail Environments
A. Rouhi, R. Aggarwal, P. Sakurikar, A. M. Namboodiri, S. P. Reddi
Embodied AI Workshop, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
2026
SABER: A Scalable Action-Based Embodied Dataset for Real-World VLA Adaptation
N. Menga, P. Sakurikar, A. Rouhi, R. Aggarwal, A. M. Namboodiri, S. P. Reddi
2026
PRISM: A Multi-View Multi-Capability Retail Video Dataset for Embodied Vision-Language Models
A. Rouhi, P. Sakurikar, S. S. Reddy, N. Menga, A. Govil, S. H. Chittajallu, R. Aggarwal, A. M. Namboodiri, S. Reddi
March 2026
Patents
Representative grantsThe Alia optical design sits on a patent family of more than 32 filings covering optics, electronics, software, and calibration. Three granted patents are listed here.
US 11,677,921
System and Method for Generating Dewarped Image Using Projection Patterns Captured from Omni-Directional Stereo Cameras
R. Aggarwal, A. M. Namboodiri, P. Sakurikar · Granted 2023 · DreamVu, Inc.
US 11,025,888
System and Method for Capturing Omni-Stereo Videos Using Multi-Sensors
R. Aggarwal, A. M. Namboodiri · Granted 2021 · DreamVu, Inc.
US 10,154,249
System and Method for Capturing Horizontal Disparity Stereo Panorama
A. M. Namboodiri, R. Aggarwal, A. Vohra · Granted 2018
Selected prior research
26 papersWork by members of the DreamVu team published before and alongside the company, across computer vision, 3D reconstruction, rendering, and depth estimation.
2021
Fast Analytic Soft Shadows from Area Lights
Aakash K. T., P. Sakurikar, P. J. Narayanan · Eurographics Symposium on Rendering (EGSR)
2021
Appearance Editing with Free-viewpoint Neural Rendering
P. Gera, Aakash K. T., D. Sirikonda, P. Sakurikar, P. J. Narayanan · arXiv preprint
2020
Understanding Dynamic Scenes Using Graph Convolution Networks
S. Mylavarapu, M. Sandhu, P. Vijayan, K. M. Krishna, B. Ravindran, A. Namboodiri · IROS
2020
Geometric Scene Refocusing
P. Sakurikar, P. J. Narayanan · arXiv preprint
2019
IDD: A Dataset for Exploring Problems of Autonomous Navigation in Unconstrained Environments
G. Varma, A. Subramanian, A. Namboodiri, M. Chandraker, C. V. Jawahar · WACV
2019
Universal Material Translator: Towards Spoof Fingerprint Generalization
R. Gajawada, A. Popli, T. Chugh, A. Namboodiri, A. K. Jain · International Conference on Biometrics (ICB)
2019
Defocus Magnification Using Conditional Adversarial Networks
P. Sakurikar, I. Mehta, P. J. Narayanan · WACV
2019
A Flexible Neural Renderer for Material Visualization
Aakash K. T., P. Sakurikar, S. Saini, P. J. Narayanan · SIGGRAPH Asia Technical Briefs
2018
Semi-Adversarial Networks: Convolutional Autoencoders for Imparting Privacy to Face Images
V. Mirjalili, S. Raschka, A. Namboodiri, A. Ross · ISBA
2018
Deep Expander Networks: Efficient Deep Networks from Graph Theory
A. Prabhu, G. Varma, A. Namboodiri · ECCV
2018
RefocusGAN: Scene Refocusing Using a Single Image
P. Sakurikar, I. Mehta, V. N. Balasubramanian, P. J. Narayanan · ECCV
2018
Structured Adversarial Training for Unsupervised Monocular Depth Estimation
I. Mehta, P. Sakurikar, P. J. Narayanan · International Conference on 3D Vision (3DV)
2018
SLFT: A Physically Accurate Framework for Tracing Synthetic Light Fields
U. Khurana, P. Sakurikar, P. J. Narayanan · ICVGIP
2017
Composite Focus Measure for High Quality Depth Maps
P. Sakurikar, P. J. Narayanan · ICCV
2017
Pose-Aware Person Recognition
V. Kumar, A. Namboodiri, M. Paluri, C. V. Jawahar · CVPR
2017
Multiscale Two-view Stereo Using Convolutional Neural Networks for Unrectified Images
P. Yadati, A. Namboodiri · MVA
2017
Recovering the 3D Geometry of Heritage Monuments from Image Collections
R. Shah, A. Deshpande, A. Namboodiri, P. J. Narayanan · Digital Hampi, Springer
2016
Panoramic Stereo Videos with a Single Camera — the Alia origin paper
R. Aggarwal, A. Vohra, A. M. Namboodiri · CVPR
2016
Detection and Segmentation of Mirror-like Surfaces Using Structured Illumination
R. Aggarwal, A. M. Namboodiri · ICVGIP
2015
Visual Phrases for Exemplar Face Detection
V. Kumar, A. Namboodiri, C. V. Jawahar · ICCV
2015
Online Handwriting Recognition Using Depth Sensors
R. Aggarwal, S. Swetha, A. M. Namboodiri, J. Sivaswamy, C. V. Jawahar · ICDAR
2010
Blind Authentication: A Secure Crypto-Biometric Verification Protocol
M. Upmanyu, A. M. Namboodiri, K. Srinathan, C. V. Jawahar · IEEE TIFS
2009
Efficient Privacy Preserving Video Surveillance
M. Upmanyu, A. M. Namboodiri, K. Srinathan, C. V. Jawahar · ICCV
2009
Contextual Restoration of Severely Degraded Document Images
J. Banerjee, A. M. Namboodiri, C. V. Jawahar · CVPR
2009
Efficient Biometric Verification in Encrypted Domain
M. Upmanyu, A. M. Namboodiri, K. Srinathan, C. V. Jawahar · ICB
2004
Online Handwritten Script Recognition
A. M. Namboodiri, A. K. Jain · IEEE TPAMI
Tell us what your model needs to learn.
Capture programs, research collaboration, and dataset partnerships.