Siddhartha Chaudhuri


Senior Research Scientist
Creative Intelligence Lab
Adobe Research, New York

Email: sidch [AT] adobe [DOT] com


Talk schedule:
Nov 2 '22: IIT Bombay
Jul 27 '22: ACM Shape Modelling Summer School
Jun 3 '22: CIPHER Seminar on Facial Recognition
Nov 15 '21: SMI Keynote
Apr 21 '21: Toronto Geometry Colloquium
Apr 14 '21: 3DGV Seminar
Jun 19 '20: CVPR Geometric Deep Learning Workshop
May 26 '20: Eurographics STAR
Mar 9 '20: IISc Computer Vision Guest Lecture
Mar 4 '20: IITB Understanding Visual Appearance Workshop
More...


I'm a computer scientist, specializing in digital geometry processing and the use of machine learning for data-driven solutions to geometric problems. My research focuses on richer tools for designing three-dimensional objects, particularly by novice and casual users, and on related problems in 3D synthesis and editing. This research is driven by a more abstract interest in shape understanding at both the structural and semantic levels. (Why is a model of a chair not one of a chicken? What distinguishes a well-designed chair from a badly designed one? How can one build a better chair?) By building computational models of form, attribute and function, we develop new techniques for designing shapes (e.g. for 3D printing or architecture), for recognizing and reconstructing objects (e.g. in scanned cities or indoor environments), and for analysing large collections of three-dimensional data such as the 3D Warehouse. Notably, most methods we explore are data-driven, since design principles and geometric characteristics are often difficult to write down by hand, and better inferred from (fully or weakly) annotated datasets. Our work combines ideas from graphics, vision, machine learning and HCI, and we have published in all of these areas.

I'm a Senior Research Scientist in the Creative Intelligence Lab at Adobe Research, New York. I received the Indian National Science Academy's Medal for Young Scientists, 2022, largely on the basis of my work while at Adobe Bangalore till mid-2023. I was an Assistant Professor (full time 2015-17, on leave 2017-22) in the department of Computer Science and Engineering at IIT Bombay, where I held an Institute Chair and received the Early Research Achiever Award. Earlier, I spent a year as a lecturer in the Computer Science department at Cornell University in the beautiful Finger Lakes region. Even earlier, I was a postdoc at Princeton University, working with Tom Funkhouser. I received my PhD from Stanford University in 2011, supervised by Vladlen Koltun. In a previous quick dip into industry, I wrote the first version of the Mixamo Fuse (later Adobe Fuse) character modeler, which used my core code upto its last released version.

[CV]  [Google Scholar]


Note: Adobe Research offers summer internships to outstanding students. Please follow the official procedure and/or talk to us when Adobe visits your university for recruiting, don't send me email. If you're interested in working with me, mention my name in your application. I do not take on students without a strong background in graphics/vision/machine learning (typically at least 2 out of 3).


News


Publications

2023

Arman Maesumi, Paul Guerrero, Vladimir Kim, Matthew Fisher, Siddhartha Chaudhuri, Noam Aigerman and Daniel Ritchie, "Explorable Mesh Deformation Subspaces from Unstructured 3D Generative Models", ACM Transactions on Graphics 42(6) (SIGGRAPH Asia conference track), 2023.
[arXiv]
Otman Benchekroun, Jiayi Eris Zhang, Siddhartha Chaudhuri, Eitan Grinspun, Yi Zhou and Alec Jacobson, "Fast Complementary Dynamics via Skinning Eigenmodes", ACM Transactions on Graphics 42(4) (SIGGRAPH journal track), 2023.
[arXiv]
Xianghao Xu, Paul Guerrero, Matthew Fisher, Siddhartha Chaudhuri and Daniel Ritchie, "Unsupervised 3D Shape Reconstruction by Part Retrieval and Assembly", CVPR, 2023.
[arXiv]

2022

Bo Sun, Vladimir Kim, Qixing Huang, Noam Aigerman and Siddhartha Chaudhuri, "PatchRD: Detail-Preserving Shape Completion by Learning Patch Retrieval and Deformation", ECCV, 2022.
[arXiv]
Kai Wang, Paul Guerrero, Vladimir Kim, Siddhartha Chaudhuri, Minhyuk Sung and Daniel Ritchie, "The Shape Part Slot Machine: Contact-based Reasoning for Generating 3D Shapes from Parts", ECCV, 2022.
[arXiv]
Noam Aigerman, Kunal Gupta, Vladimir Kim, Siddhartha Chaudhuri, Jun Saito and Thibault Groueix, "Neural Jacobian Fields: Learning Intrinsic Mappings of Arbitrary Meshes", ACM Transactions on Graphics 41(4) (SIGGRAPH journal track), 2022.
[arXiv]
Sanjeev Muralikrishnan, Siddhartha Chaudhuri, Noam Aigerman, Vladimir Kim, Matthew Fisher and Niloy Mitra, "GLASS: Geometric Latent Augmentation for Shape Spaces", CVPR, 2022.
[arXiv]

2021

Danyong Zhao, Yijing Li, Siddhartha Chaudhuri, Timothy Langlois and Jernej Barbic, "ERGOBOSS: Ergonomic Optimization of Body-Supporting Surfaces", TVCG, 2022 (to appear).
[preprint]
Pratheba Selvaraju, Mohamed Nabail, Marios Loizou, Maria Maslioukova, Melinos Averkiou, Siddhartha Chaudhuri and Evangelos Kalogerakis, "BuildingNet: Learning to Label 3D Buildings", ICCV (oral), 2021.
[project page] [preprint]
Jan Bednarik, Vladimir Kim, Siddhartha Chaudhuri, Shaifali Parashar, Mathieu Salzmann, Pascal Fua and Noam Aigerman, "Temporally-Coherent Surface Reconstruction via Metric-Consistent Atlases", ICCV, 2021.
[preprint] [video] [arXiv]
Priyadarshini K, Siddhartha Chaudhuri, Vivek Borkar and Subhasis Chaudhuri, "A Unified Batch Selection Policy for Active Metric Learning", ECML PKDD, 2021.
[preprint] [arXiv]
Zhiqin Chen, Vladimir Kim, Matthew Fisher, Noam Aigerman, Hao (Richard) Zhang and Siddhartha Chaudhuri, "DECOR-GAN: 3D Shape Detailization by Conditional Refinement", CVPR (oral), 2021.
[preprint] [code and data] [arXiv]
Mikaela Angelina Uy, Vladimir Kim, Minhyuk Sung, Noam Aigerman, Siddhartha Chaudhuri and Leonidas Guibas, "Joint Learning of 3D Shape Retrieval and Deformation", CVPR, 2021.
[project page] [preprint] [arXiv]

2020

Kangxue Yin, Zhiqin Chen, Siddhartha Chaudhuri, Matthew Fisher, Vladimir Kim and Hao (Richard) Zhang, "COALESCE: Component Assembly by Learning to Synthesize Connections", 3DV (oral), 2020.
[preprint] [supplementary] [arXiv]
Xianghao Xu, David Charatan, Sonia Raychaudhuri, Hanxiao Jiang, Mae Heitmann, Vladimir Kim, Siddhartha Chaudhuri, Manolis Savva, Angel Chang and Daniel Ritchie, "Motion Annotation Programs: A Scalable Approach to Annotating Kinematic Articulations in Large 3D Shape Collections", 3DV, 2020.
[project page] [preprint] [short video]
Gopal Sharma, Difan Liu, Evangelos Kalogerakis, Subhransu Maji, Siddhartha Chaudhuri and Radomir Mech, "ParSeNet: A Parametric Surface Fitting Network for 3D Point Clouds", ECCV, 2020.
[project page] [preprint] [arXiv]
Hsueh-Ti (Derek) Liu, Vladimir Kim, Siddhartha Chaudhuri, Noam Aigerman and Alec Jacobson, "Neural Subdivision", ACM Transactions on Graphics 39(4) (SIGGRAPH), 2020.
[project page] [preprint] [arXiv]
Priyadarshini K, Ritesh Goru, Siddhartha Chaudhuri and Subhasis Chaudhuri, "Batch Decorrelation for Active Metric Learning", IJCAI-PRICAI, 2020.
[preprint] [arXiv]
Wang Yifan, Noam Aigerman, Vladimir Kim, Siddhartha Chaudhuri and Olga Sorkine-Hornung, "Neural Cages for Detail-Preserving 3D Deformations", CVPR (oral), 2020.
[project page] [preprint] [supplementary] [arXiv]
Chenyang Zhu, Kai Xu, Siddhartha Chaudhuri, Li Yi, Leonidas Guibas and Hao (Richard) Zhang, "AdaCoSeg: Adaptive Shape Co-Segmentation with Group Consistency Loss", CVPR (oral), 2020.
[preprint] [arXiv]
Chu Wang, Babak Samari, Vladimir Kim, Siddhartha Chaudhuri and Kaleem Siddiqi, "Affinity Graph Supervision for Visual Recognition", CVPR, 2020.
[preprint] [arXiv]
Siddhartha Chaudhuri, Daniel Ritchie, Jiajun Wu, Kai Xu and Hao (Richard) Zhang, "Learning Generative Models of 3D Structures", Eurographics State-of-the-Art Reports (STAR), 2020.
[preprint]

2019

Zhiqin Chen, Kangxue Yin, Matthew Fisher, Siddhartha Chaudhuri and Hao (Richard) Zhang, "BAE-NET: Branched Autoencoder for Shape Co-Segmentation", ICCV, 2019.
[preprint] [supplementary] [code and data] [arXiv]
Priyadarshini K, Siddhartha Chaudhuri and Subhasis Chaudhuri, "PerceptNet: Learning Perceptual Similarity of Haptic Textures in Presence of Unorderable Triplets", World Haptics Conference (WHC), 2019.
[preprint] [arXiv]
Sanjeev Muralikrishnan, Vladimir Kim, Matthew Fisher and Siddhartha Chaudhuri, "Shape Unicode: A Unified Shape Representation", CVPR, 2019.
[preprint]
Manyi Li, Akshay Gadi Patil, Kai Xu, Siddhartha Chaudhuri, Owais Khan, Ariel Shamir, Changhe Tu, Baoquan Chen, Daniel Cohen-Or and Hao (Richard) Zhang, "GRAINS: Generative Recursive Autoencoders for INdoor Scenes", ACM Transactions on Graphics (TOG) 38(2), 2019.
[project page] [preprint] [arXiv]

2018

Chenyang Zhu, Kai Xu, Siddhartha Chaudhuri, Renjiao Yi and Hao (Richard) Zhang, "SCORES: Shape Composition with Recursive Substructure Priors", ACM Transactions on Graphics 37(6) (SIGGRAPH Asia), 2018.
[preprint] [arXiv]
Hubert Lin, Melinos Averkiou, Evangelos Kalogerakis, Balazs Kovacs, Siddhant Ranade, Vladimir Kim, Siddhartha Chaudhuri and Kavita Bala, "Learning Material-Aware Local Descriptors for 3D Shapes", 3DV, 2018.
[preprint] [supplementary] [data] [arXiv]
Sanjeev Muralikrishnan, Vladimir Kim and Siddhartha Chaudhuri, "Tags2Parts: Discovering Semantic Regions from Shape Tags", CVPR, 2018.
[project page] [preprint] [arXiv]
Shiv Shankar, Vihari Piratla, Soumen Chakrabarti, Siddhartha Chaudhuri, Preethi Jyothi and Sunita Sarawagi, "Generalizing Across Domains via Cross-Gradient Training", ICLR, 2018.
[preprint] [OpenReview]
Haibin Huang, Evangelos Kalogerakis, Siddhartha Chaudhuri, Duygu Ceylan, Vladimir Kim and Ersin Yumer, "Learning Local Shape Descriptors from Part Correspondences with Multi-view Convolutional Networks", ACM Transactions on Graphics (TOG) 37(1), 2018.
[project page] [preprint] [arXiv]

2017

Minhyuk Sung, Hao Su, Vladimir Kim, Siddhartha Chaudhuri and Leonidas Guibas, "ComplementMe: Weakly-Supervised Component Suggestion for 3D Modeling", ACM Transactions on Graphics 36(6) (SIGGRAPH Asia), 2017.
[project page] [preprint] [press release] [arXiv]
Jun Li, Kai Xu, Siddhartha Chaudhuri, Ersin Yumer, Hao (Richard) Zhang and Leonidas Guibas, "GRASS: Generative Recursive Autoencoders for Shape Structures", ACM Transactions on Graphics 36(4) (SIGGRAPH), 2017.
[project page] [preprint] [press release]
Evangelos Kalogerakis, Melinos Averkiou, Subhransu Maji and Siddhartha Chaudhuri, "3D Shape Segmentation with Projective Convolutional Networks", CVPR (oral), 2017.
[project page] [preprint]
Utkarsh Mall, G. Roshan Lal, Siddhartha Chaudhuri and Parag Chaudhuri, "A Deep Recurrent Framework for Cleaning Motion Capture Data", CoRR (not peer-reviewed), 2017.
[arXiv]

2016 and earlier

Xuekun Guo, Juncong Lin, Kai Xu, Siddhartha Chaudhuri and Xiaogang Jin, "CustomCut: On-demand Extraction of Customized 3D Parts with 2D Sketches", Symposium on Geometry Processing (SGP), 2016.
[preprint]
Ersin Yumer, Siddhartha Chaudhuri, Jessica Hodgins and Levent Burak Kara, "Semantic Shape Editing Using Deformation Handles", ACM Transactions on Graphics 34(4) (SIGGRAPH), 2015.
[preprint] [supplementary]
Tianqiang Liu, Siddhartha Chaudhuri, Vladimir Kim, Qi-Xing Huang, Niloy Mitra and Thomas Funkhouser, "Creating Consistent Scene Graphs Using a Probabilistic Grammar", ACM Transactions on Graphics 33(6) (SIGGRAPH Asia), 2014.
[project page] [preprint]
Vladimir Kim, Siddhartha Chaudhuri, Leonidas Guibas and Thomas Funkhouser, "Shape2Pose: Human-Centric Shape Analysis", ACM Transactions on Graphics 33(4) (SIGGRAPH), 2014.
[project page] [preprint]
Siddhartha Chaudhuri, Evangelos Kalogerakis, Stephen Giguere and Thomas Funkhouser, "AttribIt: Content Creation with Semantic Attributes", UIST, 2013.
[project page] [preprint] [video (50 MB)]
Vladimir Kim, Wilmot Li, Niloy Mitra, Siddhartha Chaudhuri, Stephen DiVerdi and Thomas Funkhouser, "Learning Part-Based Templates from Large Collections of 3D Shapes", ACM Transactions on Graphics 32(4) (SIGGRAPH), 2013.
[project page] [preprint]
Evangelos Kalogerakis, Siddhartha Chaudhuri, Daphne Koller and Vladlen Koltun, "A Probabilistic Model for Component-Based Shape Synthesis", ACM Transactions on Graphics 31(4) (SIGGRAPH), 2012.
[project page] [preprint] [video (112 MB)]
Siddhartha Chaudhuri, Evangelos Kalogerakis, Leonidas Guibas and Vladlen Koltun, "Probabilistic Reasoning for Assembly-Based 3D Modeling", ACM Transactions on Graphics 30(4) (SIGGRAPH), 2011.
[project page] [preprint] [video (44 MB)]
Siddhartha Chaudhuri and Vladlen Koltun, "Data-Driven Suggestions for Creativity Support in 3D Modeling", ACM Transactions on Graphics 29(6) (SIGGRAPH Asia), 2010.
[journal] [preprint] [video (58 MB)]
Siddhartha Chaudhuri and Vladlen Koltun, "Smoothed Analysis of Probabilistic Roadmaps", Computational Geometry: Theory and Applications 42(8), pp. 731-747, 2009.
[journal] [preprint]
Siddhartha Chaudhuri, Daniel Horn, Pat Hanrahan and Vladlen Koltun, "Image-Based Exploration of Massive Online Environments", Stanford University Computer Science Technical Report (not peer-reviewed), CSTR 2009-02, 2009.
[pdf]
Siddhartha Chaudhuri, Randhir K. Singh and Edoardo Charbon, "Feature-Based Techniques for Real-Time Morphable Model Facial Image Analysis", Image and Video Communications and Processing Conference, IS&T/SPIE's 17th Annual Symposium on Electronic Imaging Science and Technology, San Jose, 2005.
[pdf] [extended version]
Siddhartha Chaudhuri, Ratan K. Ghosh and Sajal K. Das, "Towards Optimal Sensor Placement with Hypercube Cutting Planes", IEEE Wireless Communications and Networking Conference (invited paper), New Orleans, 2005.
[pdf] [extended version]
Manu Chhabra et al, "Novel Approaches to Vision and Motion Control for Robot Soccer", National Conference on Advanced Manufacturing and Robotics, CMERI, Durgapur, 2004.
[pdf]

Teaching


Code


I wrote and maintain Thea, a freely available, BSD-licensed library of C++ classes for computer graphics, primarily for 3D geometry processing (not to be confused with the concurrently developed Thea Render, which I am not connected with). It is the core library I use for nearly all my research projects, and it is also the core library for Adobe Fuse, which I originally authored.


Links