Mockup
Users can apply their 2D designs into real photographs in a 3D-aware manner.
Matheus Gadelha
I am currently a Senior Research Scientist at Adobe Research. I received my PhD from University of Massachusetts - Amherst while being supervised by Prof. Rui Wang and Prof. Subhransu Maji. I am interested in Computer Graphics, Vision and their intersections with Machine Learning. My work is focused on models and representations of tridimensional data for both discriminative and generative models.
Lately, I have been particularly interested in mechanisms to incorporate 3D capabilities into large generative models so we can properly control them and robustly create/understand 3D data. I am also broadly interested in techniques (not necessarily ML-based) that allow us to better manipulate and author 3D content.
Internships for PhD students: If you are interested in related areas to the ones I've mentioned above (or anything related to my previous research), don't be shy and send me an e-mail with your CV and a short description of the problems you are interested in working on. We are always looking for talented interns to join us at Adobe Research.
Pro-bono Mentoring: If you are a student/early-career researcher/enthusiast seeking advice on research projects or general career directions, you can use this link to schedule a chat with me. I am particularly interested in chatting with folks from South/Latin America and helping in any way I can.
Academic Collaborations: If you are a professor or a student interested in working with me, feel free to send me an e-mail -- I am more than happy to discuss interesting research problems we could work on together (or even just to chat about research).
[CV] [bluesky] [linkedin] [email] [scholar]
Research Interns
I have been incredibly fortunate to have worked with amazing students. I keep a list of all the interns I have mentored/co-mentored below, and I am always happy to talk to students interested in research. Feel free to reach out!- Umangi Jain, University of Toronto
- Rio Aguina-Kang, UC San Diego
now at Drafted - Sanghyun Son, University of Maryland
now at Genesis AI - Edurne Bernal-Berdun, Universidad de Zaragoza
- Dima Petrov, UMass Amherst
now at Foundation AGI - Marissa Ramirez de Chanlatte, UC Berkeley
now at Playground Global
- Aditya Ganeshan, Brown University
- Dale Decatur, University of Chicago
- Karran Pandey, University of Toronto
- Shengqu Cai, ETH Zürich
now at Stanford University - Ta-Ying Cheng, Oxford
now at Netflix
- Oindrila Saha, UMass Amherst
now at Adobe Research - Yen-Chi Cheng, UIUC
- Amir Barda, Tel Aviv University
- Leticia Mattos, MIT
- Yun-Chun Chen, University of Toronto
now at Adobe Firefly Foundry
Software
*I did not develop these by myself, but the list of collaborators is too big to include here. They are the result of a big team effort.
3D to Image
Using image generative models to "render" a 3D scene according to a text prompt and style presets.
3D to Image
Easily create 3D shapes using Project Neo and render them using a text-to-image generative model.
Papers
ViPS: Video-informed Pose Spaces for Auto-Rigged Meshes
Honglin Chen, Karran Pandey, Rundi Wu, Matheus Gadelha, Yannick Hold-Geoffroy, Ayush Tewari, Niloy J. Mitra, Changxi Zheng, Paul Guerrero
GimbalDiffusion: Gravity-Aware Camera Control for Video Generation
Frédéric Fortier-Chouinard, Yannick Hold-Geoffroy, Valentin Deschaintre, Matheus Gadelha, Jean-François Lalonde
Material Magic Wand: Material-Aware Grouping of 3D Parts in Untextured Meshes
Umangi Jain, Vladimir Kim, Matheus Gadelha, Igor Gilitschenski, Zhiqin Chen
Residual Primitive Fitting of 3D Shapes with SuperFrusta
Aditya Ganeshan, Matheus Gadelha, Thibault Groueix, Zhiqin Chen, Siddhartha Chaudhuri, Vladimir Kim, Wang Yifan, Daniel Ritchie
3D Space as a Scratchpad for Editable Text-to-Image Generation
Oindrila Saha, Vojtech Krs, Radomir Mech, Subhransu Maji, Matheus Gadelha*, Kevin Blackburn-Matzen*
MeshSplatting: Differentiable Rendering with Opaque Meshes
Jan Held, Sanghyun Son, Renaud Vandeghen, Daniel Rebain, Matheus Gadelha, Yi Zhou, Anthony Cioppa, Ming C. Lin, Marc Van Droogenbroeck, Andrea Tagliasacchi
SIGMA-Gen: Structure and Identity Guided Multi-subject Assembly for Image Generation
Oindrila Saha, Vojtech Krs, Radomir Mech, Subhransu Maji, Kevin Blackburn-Matzen*, Matheus Gadelha*
Seeing Through Clutter: Structured 3D Scene Reconstruction via Iterative Object Removal
Rio Aguina-Kang, Kevin Blackburn-Matzen, Thibault Groueix, Vladimir Kim, Matheus Gadelha
DepthScape: Authoring 2.5D Designs via Depth Estimation, Semantic Understanding, and Geometry Extraction
Xia Su, Cuong Nguyen, Matheus Gadelha, Jon E. Froehlich
Frame In-N-Out: Unbounded Controllable Image-to-Video Generation
Boyang Wang, Xuweiyi Chen, Matheus Gadelha, Zezhou Cheng
Reusing Computation in Text‑to‑Image Diffusion for Efficient Generation of Image Sets
Dale Decatur, Thibault Groueix, Yifan Wang, Rana Hanocka, Vova Kim, Matheus Gadelha
DMesh++: An Efficient Differentiable Mesh for Complex Shapes
Sanghyun Son, Matheus Gadelha, Yang Zhou, Matthew Fisher, Zexiang Xu, Ming C. Lin, Yi Zhou
3D-Fixup: Advancing Photo Editing with 3D Priors
Yen-Chi Cheng, Krishna Kumar Singh, Jae Shin Yoon, Alex Schwing, Liangyan Gui, Matheus Gadelha, Paul Guerrero, Nanxuan Zhao
PreciseCam: Precise Camera Control for Text-to-Image Generation
Edurne Bernal-Berdun, Ana Serrano, Belen Masia, Matheus Gadelha, Yannick Hold-Geoffroy, Xin Sun, Diego Gutierrez
Motion Modes: What Could Happen Next?
Karran Pandey, Matheus Gadelha, Yannick Hold-Geoffroy, Karan Singh, Niloy J. Mitra, Paul Guerrero
Instant3dit: Multiview Inpainting for Fast Editing of 3D Objects
Amir Barda, Matheus Gadelha, Vladimir Kim, Noam Aigerman, Amit Haim Bermano, Thibault Groueix
Text-guided Controllable Mesh Refinement for Interactive 3D Modeling
Yun-Chun Chen, Selena Ling, Zhiqin Chen, Vladimir G. Kim, Matheus Gadelha, Alec Jacobson
DMesh: A Differentiable Representation for General Meshes
Sanghyun Son, Matheus Gadelha, Yang Zhou, Zexiang Xu, Ming C. Lin, Yi Zhou
GEM3D: Generative Medial Abstractions for 3D Shape Synthesis
Dmitry Petrov, Pradyumn Goyal, Vikas Thamizharasan, Vova Kim, Matheus Gadelha, Melinos Averkiou, Siddhartha Chaudhuri, Evangelos Kalogerakis
Learning Continuous 3D Words for Text-to-Image Generation
Ta-Ying Cheng, Matheus Gadelha, Thibault Groueix, Matthew Fisher, Radomir Mech, Andrew Markham, Niki Trigoni
Diffusion Handles: Enabling 3D Edits for Diffusion Models by Lifting Activations to 3D
Karran Pandey, Paul Guerrero, Matheus Gadelha, Yannick Hold-Geoffroy, Karan Singh, Niloy Mitra
Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models
Shengqu Cai, Duygu Ceylan, Matheus Gadelha, Chun-Hao Huang, Tuanfeng Y. Wang, Gordon Wetzstein
3DMiner: Discovering Shapes from Large-Scale Unannotated Image Datasets
Ta-Ying Cheng, Matheus Gadelha, Soren Pirk, Thibault Groueix, Radomir Mech, Andrew Markham, Niki Trigoni
ANISE: Assembly-based Neural Implicit Surface rEconstruction
Dmitry Petrov, Matheus Gadelha, Radomir Mech, Evangelos Kalogerakis
Accidental Turntables: Learning 3D Pose by Watching Objects Turn
Zezhou Cheng, Matheus Gadelha, Subhransu Maji
Recovering Detail in 3D Shapes Using Disparity Maps
Marissa Ramirez de Chanlatte, Matheus Gadelha, Thibault Groueix, Radomir Mech
PrimFit: Learning to Fit Primitives Improves Few Shot Learning on Point Clouds
Gopal Sharma, Bidya Dash, Matheus Gadelha, Aruni RoyChowdhury, Marios Loizou, Evangelos Kalogerakis, Liangliang Cao, Erik Learned-Miller, Rui Wang and Subhransu Maji
PlanarRecon: Real-time 3D Plane Detection and Reconstruction from Posed Monocular Videos
Yiming Xie, Matheus Gadelha, Fengting Yang, Xiaowei Zhou, Huaizu Jiang
Trace Match & Merge: Long-TermField-Of-View Prediction for AR Applications
Adam Viola*, Sahil Sharma*,Pankaj Bishnoi*, Matheus Gadelha, Stefano Petrangeli, Haoliang Wang, Viswanathan Swaminathan
Best paper candidate at IEEE AIVR.
Inferring 3D Shapes from Image Collections using Adversarial Networks
Matheus Gadelha, Aartika Rai, Rui Wang, Subhransu Maji
Label-Efficient Learning on Point Clouds using Approximate Convex Decompositions
Matheus Gadelha*, Aruni RoyChowdhury*, Gopal Sharma, Evangelos Kalogerakis, Liangliang Cao, Erik Learned-Miller, Rui Wang, Subhransu Maji
Deep Manifold Prior
Matheus Gadelha, Rui Wang, Subhransu Maji
Best poster honorable mention at NECV.
Learning Generative Models of Shape Handles
Matheus Gadelha, Giorgio Gori, Duygu Ceylan, Radomir Mech, Nathan Carr, Tamy Boubekeur, Rui Wang, Subhransu Maji
Shape Reconstruction with Differentiable Projections and Deep Priors
Matheus Gadelha, Rui Wang, Subhransu Maji
A Bayesian Perspective on the Deep Image Prior
Zezhou Cheng, Matheus Gadelha, Subhransu Maji, Daniel Sheldon
A Deeper Look at 3D Shape Classifiers
Jong-Chyi Su, Matheus Gadelha, Rui Wang, Subhransu Maji
Multiresolution Tree Networks for Point Cloud Processing
Matheus Gadelha, Rui Wang, Subhransu Maji
Unsupervised 3D Shape Induction from 2D Views of Multiple Objects
Matheus Gadelha, Subhransu Maji, Rui Wang
3D Shape Reconstruction from Sketches via Multi-view Convolutional Networks
Zhaoliang Lun, Matheus Gadelha, Evangelos Kalogerakis, Subhransu Maji, Rui Wang
Shape Generation using Spatially Partitioned Point Clouds
Matheus Gadelha, Subhransu Maji, Rui Wang
Service
I worked/am working as a reviewer for the following venues:
- SIGGRAPH 2023-2025
- CVPR 2018-2024 (2021 outstanding reviewer)
- ICCV 2019-2023
- ECCV 2018-2023
- ICLR 2021
- TPAMI (2018, 2021, 2023)
- IJCV 2023, 2024
- Computer and Graphics 2018, 2024
- SIGGRAPH Asia 2018, 2024
- Pacific Graphics 2019
I worked/am working as AC for the following venues:
- WACV 2024, 2025
- CVPR 2025, 2026
Professional Experience
Adobe.
Research Scientist (Seattle, WA).
June 2021 - Now.
Google (Perception).
Research Intern (Amherst, MA).
Summer 2020.
Adobe.
Research Intern (San Jose, CA)
Summer 2019.
Amazon.
Applied Scientist Intern (Pasadena, CA)
Summer 2018.
Federal University at Rio Grande do Norte.
Temporary Letcturer - Introduction to Algorithms and Numerical Methods.
June 2014 - June2015 .