Few-Shot Generalization for Single-Image 3D Reconstruction via Priors
By Bram Wallace, Bharath Hariharan. 2019
45 citations recorded by the source corpus.
Recent work on single-view 3D reconstruction shows impressive results, but has been restricted to a few fixed categories where extensive training data is available. The problem of generalizing these models to new classes with limited training data is largely open. To address this problem, we present a new model architecture that reframes single-view 3D reconstruction as learnt, category agnostic refinement of a provided, category-specific prior. The provided prior shape for a novel class can be obtained from as few as one 3D shape from this class. Our model can start reconstructing objects from the novel class using this prior without seeing any training image for this class and without any retraining. Our model outperforms category-agnostic baselines and remains competitive with more sophisticated baselines that finetune on the novel categories. Additionally, our network is capable of improving the reconstruction given multiple views despite not being trained on task of multi-view reconstruction.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, 3D Vision & Pose, Meta & Continual Learning
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Jitendra Malik
- Christoph Feichtenhofer
- Georgia Gkioxari
- Abhinav Kumar Gupta
- Justin Johnson
- Ross B. Girshick
- Cordelia Schmid
- Trevor Darrell
In Computer Vision and Pattern Recognition
- Jitendra Malik
- Luc Van Gool
- Christoph Feichtenhofer
- Trevor Darrell
- Xiaogang Wang
- Cordelia Schmid
- Andrew Zisserman
- Dacheng Tao
In 3D Vision & Pose
- Georgia Gkioxari
- Justin Johnson
- Jitendra Malik
- Christoph Feichtenhofer
- Chaozheng Wu
- Meng Tian
- Marcelo H. Ang Jr.
- Gim Hee Lee
In Meta & Continual Learning
Citations in the corpus
- Multiview Compressive Coding for 3D Reconstruction (2023, background): “Some [66, 77] explore category-agnostic models, but focus on synthetic datasets.”
- Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation (2020, methodology): “Unlike methods such as simple averaging [30] and principal component analysis (PCA) [3] that operate on voxel representations, our autoencoder framework can be easily altered to take any 3D representations.”