On Extending Semantic Abstraction for Efficient Search of Hidden Objects
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911340512149504 |
|---|---|
| author | Pais, Tasha Belulkar, Nikhilesh |
| author_facet | Pais, Tasha Belulkar, Nikhilesh |
| contents | Semantic Abstraction's key observation is that 2D VLMs' relevancy activations roughly correspond to their confidence of whether and where an object is in the scene. Thus, relevancy maps are treated as "abstract object" representations. We use this framework for learning 3D localization and completion for the exclusive domain of hidden objects, defined as objects that cannot be directly identified by a VLM because they are at least partially occluded. This process of localizing hidden objects is a form of unstructured search that can be performed more efficiently using historical data of where an object is frequently placed. Our model can accurately identify the complete 3D location of a hidden object on the first try significantly faster than a naive random search. These extensions to semantic abstraction hope to provide household robots with the skills necessary to save time and effort when looking for lost objects. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_22220 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On Extending Semantic Abstraction for Efficient Search of Hidden Objects Pais, Tasha Belulkar, Nikhilesh Computer Vision and Pattern Recognition Artificial Intelligence Robotics Semantic Abstraction's key observation is that 2D VLMs' relevancy activations roughly correspond to their confidence of whether and where an object is in the scene. Thus, relevancy maps are treated as "abstract object" representations. We use this framework for learning 3D localization and completion for the exclusive domain of hidden objects, defined as objects that cannot be directly identified by a VLM because they are at least partially occluded. This process of localizing hidden objects is a form of unstructured search that can be performed more efficiently using historical data of where an object is frequently placed. Our model can accurately identify the complete 3D location of a hidden object on the first try significantly faster than a naive random search. These extensions to semantic abstraction hope to provide household robots with the skills necessary to save time and effort when looking for lost objects. |
| title | On Extending Semantic Abstraction for Efficient Search of Hidden Objects |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2512.22220 |