Robust Change Detection Based on Neural Descriptor Fields
2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) October 23-27, 2022, Kyoto, Japan
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024
|
| Online Access: | https://hdl.handle.net/1721.1/153751 |
| _version_ | 1826210622414520320 |
|---|---|
| author | Fu, Jiahui Du, Yilun Singh, Kurran Tenenbaum, Joshua B. Leonard, John J. |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Fu, Jiahui Du, Yilun Singh, Kurran Tenenbaum, Joshua B. Leonard, John J. |
| author_sort | Fu, Jiahui |
| collection | MIT |
| description | 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) October 23-27, 2022, Kyoto, Japan |
| first_indexed | 2024-09-23T14:52:33Z |
| format | Article |
| id | mit-1721.1/153751 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T14:52:33Z |
| publishDate | 2024 |
| publisher | IEEE |
| record_format | dspace |
| spelling | mit-1721.1/1537512024-09-20T18:52:12Z Robust Change Detection Based on Neural Descriptor Fields Fu, Jiahui Du, Yilun Singh, Kurran Tenenbaum, Joshua B. Leonard, John J. Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) October 23-27, 2022, Kyoto, Japan The ability to reason about changes in the environment is crucial for robots operating over extended periods of time. Agents are expected to capture changes during operation so that actions can be followed to ensure a smooth progression of the working session. However, varying viewing angles and accumulated localization errors make it easy for robots to falsely detect changes in the surrounding world due to low observation overlap and drifted object associations. In this paper, based on the recently proposed category-level Neural Descriptor Fields (NDFs), we develop an object-level online change detection approach that is robust to partially overlapping observations and noisy localization results. Utilizing the shape completion capability and SE(3)-equivariance of NDFs, we represent objects with compact shape codes encoding full object shapes from partial observations. The objects are then organized in a spatial tree structure based on object centers recovered from NDFs for fast queries of object neighborhoods. By associating objects via shape code similarity and comparing local object-neighbor spatial layout, our proposed approach demonstrates robustness to low observation overlap and localization noises. We conduct experiments on both synthetic and real-world sequences and achieve improved change detection results compared to multiple baseline methods. 2024-03-13T20:04:43Z 2024-03-13T20:04:43Z 2022-10-23 2024-03-13T19:54:59Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/153751 Fu, Jiahui, Du, Yilun, Singh, Kurran, Tenenbaum, Joshua B. and Leonard, John J. 2022. "Robust Change Detection Based on Neural Descriptor Fields." en 10.1109/iros47612.2022.9981246 Creative Commons Attribution-Noncommercial-ShareAlike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf IEEE arxiv |
| spellingShingle | Fu, Jiahui Du, Yilun Singh, Kurran Tenenbaum, Joshua B. Leonard, John J. Robust Change Detection Based on Neural Descriptor Fields |
| title | Robust Change Detection Based on Neural Descriptor Fields |
| title_full | Robust Change Detection Based on Neural Descriptor Fields |
| title_fullStr | Robust Change Detection Based on Neural Descriptor Fields |
| title_full_unstemmed | Robust Change Detection Based on Neural Descriptor Fields |
| title_short | Robust Change Detection Based on Neural Descriptor Fields |
| title_sort | robust change detection based on neural descriptor fields |
| url | https://hdl.handle.net/1721.1/153751 |
| work_keys_str_mv | AT fujiahui robustchangedetectionbasedonneuraldescriptorfields AT duyilun robustchangedetectionbasedonneuraldescriptorfields AT singhkurran robustchangedetectionbasedonneuraldescriptorfields AT tenenbaumjoshuab robustchangedetectionbasedonneuraldescriptorfields AT leonardjohnj robustchangedetectionbasedonneuraldescriptorfields |