# Object-Centric Multi-View Aggregation

arXiv: https://arxiv.org/abs/2007.10300 · HTML: https://mlsanity.com/paper/2007.10300/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jul 2020.

## Built on by (most cited first, 6 of 6)

- [Neural Kernel Surface Reconstruction](https://mlsanity.com/paper/2305.19590/) (2023)
- [Neural Fields as Learnable Kernels for 3D Reconstruction](https://mlsanity.com/paper/2111.13674/) (2021)
- [StrobeNet: Category-Level Multiview Reconstruction of Articulated Objects](https://mlsanity.com/paper/2105.08016/) (2021)
- [Robust Symmetry Detection via Riemannian Langevin Dynamics](https://mlsanity.com/paper/2410.02786/) (2024)
- [Quartet of Diffusions: Structure-Aware Point Cloud Generation through Part and Symmetry Guidance](https://mlsanity.com/paper/2601.20425/) (2026)
- [Symmetry Matters: Auditing and Symmetrizing 3D Generative Models](https://mlsanity.com/paper/2512.18953/) (2025)

## Abstract (arXiv)

We present an approach for aggregating a sparse set of views of an object in order to compute a semi-implicit 3D representation in the form of a volumetric feature grid. Key to our approach is an object-centric canonical 3D coordinate system into which views can be lifted, without explicit camera pose estimation, and then combined -- in a manner that can accommodate a variable number of views and is view order independent. We show that computing a symmetry-aware mapping from pixels to the canonical coordinate system allows us to better propagate information to unseen regions, as well as to robustly overcome pose ambiguities during inference. Our aggregate representation enables us to perform 3D inference tasks like volumetric reconstruction and novel view synthesis, and we use these tasks to demonstrate the benefits of our aggregation approach as compared to implicit or camera-centric alternatives.