RESEARCH OVERVIEW

We study computer vision and machine learning. Our primary interests include:
3D Vision: Single-view and multi-view 3D reconstruction, in particular, per-pixel reconstruction of geometry and motion for arbitrary in-the-wild scenes.
Object and Action Recognition: Understanding "what is there" (objects and their locations) as well as "what is going on" (interactions and relationships).
Automated Reasoning: Automated theorem proving and its connections to NLP, program synthesis, and AutoML.

LAB MEMBERS

Faculty / Principal Investigator: Jia Deng
PhD Students: Stamatis Alexandropoulos . Beining Han . Karhan Kayan . Erich Liang . Lahav Lipson . Zeyu Ma . Lingjie Mei . Meenal Parakh . Alex Raistrick . Yihan Wang . Hongyu Wen . Siyang Wu . Yiming Zuo
Undergraduate Students: David Yan

CONTACT

Department of Computer Science
Princeton University
35 Olden Street, Princeton, NJ 08540-5233

RECENT PAPERS

Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generation
Alexander Raistrick*, Lingjie Mei*, Karhan Kayan*, David Yan, Yiming Zuo, Beining Han, Hongyu Wen, Meenal Parakh, Stamatis Alexandropoulos, Lahav Lipson, Zeyu Ma, Jia Deng (*equal contribution)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024

Multi-Session SLAM with Differentiable Wide-Baseline Pose Optimization
Lahav Lipson, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024

Llemma: An Open Language Model For Mathematics
Zhangir Azerbayev, Hailey Schoelkopf, Keiran Paster, Marco Dos Santos, Stephen McAleer, Albert Q. Jiang, Jia Deng, Stella Biderman, Sean Welleck
International Conference on Learning Representations (ICLR), 2024
[ paper ] [ code ]

Label-Free Synthetic Pretraining of Object Detectors
Hei Law, Jia Deng
Winter Conference on Applications of Computer Vision (WACV), 2024
[ paper ] [ code ]

View-Dependent Octree-based Mesh Extraction in Unbounded Scenes for Procedural Synthetic Data
Zeyu Ma, Alexander Raistrick, Lahav Lipson, Jia Deng
arXiv:2312.08364
[ paper ] [ code ]

Deep Patch Visual Odometry
Zachary Teed*, Lahav Lipson*, Jia Deng (*equal contribution)
Neural Information Processing Systems (NeurIPS), 2023
[ paper ] [ code ]

Siamese Masked Autoencoders
Agrim Gupta, Jiajun Wu, Jia Deng, Li Fei-Fei
Neural Information Processing Systems (NeurIPS), 2023
[ paper ] [ project ]

Convolutional Networks with Oriented 1D Kernels
Alexandre Kirchmeyer, Jia Deng
International Conference on Computer Vision (ICCV), 2023
[ paper ] [ code ]

Learning Symbolic Rules for Reasoning in Quasi-Natural Language
Kaiyu Yang, Jia Deng
Transactions on Machine Learning Research (TMLR), 2023
[ paper ] [ code ]

Infinite Photorealistic Worlds using Procedural Generation
Alexander Raistrick*, Lahav Lipson*, Zeyu Ma*, Lingjie Mei, Mingzhe Wang, Yiming Zuo, Karhan Kayan, Hongyu Wen, Beining Han, Yihan Wang, Alejandro Newell, Hei Law, Ankit Goyal, Kaiyu Yang, Jia Deng (*equal contribution)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023
[ paper ] [ project ] [ code ]

View Synthesis with Sculpted Neural Points
Yiming Zuo, Jia Deng
International Conference on Learning Representations (ICLR), 2023
[ paper ] [ code ]

Generating Natural Language Proofs with Verifier-Guided Search
Kaiyu Yang, Jia Deng, Danqi Chen
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2022
[ paper ] [ code ]

Non-deep Networks
Ankit Goyal, Alexey Bochkovskiy, Jia Deng, Vladlen Koltun
Neural Information Processing Systems (NeurIPS), 2022
[ paper ] [ code ]

Multiview Stereo with Cascaded Epipolar RAFT
Zeyu Ma, Zachary Teed, Jia Deng
European Conference on Computer Vision (ECCV), 2022
[ paper ] [ code ]

A Study of Face Obfuscation in ImageNet
Kaiyu Yang, Jacqueline Yau, Li Fei-Fei, Jia Deng, Olga Russakovsky
International Conference on Machine Learning (ICML), 2022
[ paper ] [ code ]

IFOR: Iterative Flow Minimization for Robotic Object Rearrangement
Ankit Goyal, Arsalan Mousavian, Chris Paxton, Yu-Wei Chao, Brian Okorn, Jia Deng, Dieter Fox
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022
[ paper ] [ project ]

Coupled Iterative Refinement for 6D Multi-Object Pose Estimation
Lahav Lipson, Zachary Teed, Ankit Goyal, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022
[ paper ] [ code ]

RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching
Lahav Lipson, Zachary Teed, Jia Deng
International Conference on 3D Vision (3DV), 2021
[ paper ] [ code ] [ Best Student Paper Award ]

DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras
Zachary Teed, Jia Deng
Neural Information Processing Systems (NeurIPS), 2021
[ paper ] [ code ]

Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline
Ankit Goyal, Hei Law, Bowei Liu, Alejandro Newell, Jia Deng
International Conference on Machine Learning (ICML), 2021
[ paper ] [ code ]

Tangent Space Backpropagation for 3D Transformation Groups
Zachary Teed, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021
[ paper ] [ code ]

RAFT-3D: Scene Flow using Rigid-Motion Embeddings
Zachary Teed, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021
[ paper ] [ code ]

Dynamically Grown Generative Adversarial Networks
Lanlan Liu, Yuting Zhang, Jia Deng, Stefano Soattos
AAAI Conference on Artificial Intelligence (AAAI), 2021
[ paper ]

Learning to Sit: Synthesizing Human-Chair Interactions via Hierarchical Control
Yu-Wei Chao, Jimei Yang, Weifeng Chen, Jia Deng
AAAI Conference on Artificial Intelligence (AAAI), 2021
[ paper ]

Learning to Prove Theorems by Learning to Generate Theorems
Mingzhe Wang, Jia Deng
Neural Information Processing Systems (NeurIPS), 2020
[ paper ] [ code ]

Rel3D: A Minimally Contrastive Benchmark for Grounding Spatial Relations in 3D
Ankit Goyal, Kaiyu Yang, Dawei Yang, Jia Deng
Neural Information Processing Systems (NeurIPS), 2020
[ paper ] [ code ] [ data ]

Strongly Incremental Constituency Parsing with Graph Neural Networks
Kaiyu Yang, Jia Deng
Neural Information Processing Systems (NeurIPS), 2020
[ paper ] [ code ]

RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
Zachary Teed, Jia Deng
European Conference on Computer Vision (ECCV), 2020
[ paper ] [ code ] [ Best Paper Award ]

A Unified Framework of Surrogate Loss by Refactoring and Interpolation
Lanlan Liu, Mingzhe Wang, Jia Deng
European Conference on Computer Vision (ECCV), 2020
[ paper ] [ code ]

CornerNet-Lite: Efficient Keypoint Based Object Detection
Hei Law, Yun Teng, Olga Russakovsky, Jia Deng
British Machine Vision Conference (BMVC), 2020
[ paper ] [ code ]

PackIt: A Virtual Environment for Geometric Planning
Ankit Goyal, Jia Deng
International Conference on Machine Learning (ICML), 2020
[ paper ] [ code ]

How Useful is Self-Supervised Pretraining for Visual Tasks?
Alejandro Newell, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
[ paper ] [ code ]

OASIS: A Large-Scale Dataset for Single Image 3D in the Wild
Weifeng Chen, Shengyi Qian, David Fan, Noriyuki Kojima, Max Hamilton, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
[ paper ] [ project ]

Learning to Generate Synthetic 3D Training Data through Hybrid Gradient
Dawei Yang, Jia Deng
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
[ paper ]

DeepV2D: Video to Depth with Differentiable Structure from Motion
Zachary Teed, Jia Deng
International Conference on Learning Representations (ICLR), 2020
[ paper ] [ code ]

D3D: Distilled 3D Networks for Video Action Recognition
Jonathan Stroud, David A. Ross, Chen Sun, Jia Deng, Rahul Sukthankar
Winter Conference on Applications of Computer Vision (WACV), 2020
[ paper ] [ code ]


more...