SPOT: Self-Training with Patch-Order Permutation for Object-Centric Learning with Autoregressive Transformers
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911828300267520 |
|---|---|
| author | Kakogeorgiou, Ioannis Gidaris, Spyros Karantzalos, Konstantinos Komodakis, Nikos |
| author_facet | Kakogeorgiou, Ioannis Gidaris, Spyros Karantzalos, Konstantinos Komodakis, Nikos |
| contents | Unsupervised object-centric learning aims to decompose scenes into interpretable object entities, termed slots. Slot-based auto-encoders stand out as a prominent method for this task. Within them, crucial aspects include guiding the encoder to generate object-specific slots and ensuring the decoder utilizes them during reconstruction. This work introduces two novel techniques, (i) an attention-based self-training approach, which distills superior slot-based attention masks from the decoder to the encoder, enhancing object segmentation, and (ii) an innovative patch-order permutation strategy for autoregressive transformers that strengthens the role of slot vectors in reconstruction. The effectiveness of these strategies is showcased experimentally. The combined approach significantly surpasses prior slot-based autoencoder methods in unsupervised object segmentation, especially with complex real-world images. We provide the implementation code at https://github.com/gkakogeorgiou/spot . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_00648 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | SPOT: Self-Training with Patch-Order Permutation for Object-Centric Learning with Autoregressive Transformers Kakogeorgiou, Ioannis Gidaris, Spyros Karantzalos, Konstantinos Komodakis, Nikos Computer Vision and Pattern Recognition Unsupervised object-centric learning aims to decompose scenes into interpretable object entities, termed slots. Slot-based auto-encoders stand out as a prominent method for this task. Within them, crucial aspects include guiding the encoder to generate object-specific slots and ensuring the decoder utilizes them during reconstruction. This work introduces two novel techniques, (i) an attention-based self-training approach, which distills superior slot-based attention masks from the decoder to the encoder, enhancing object segmentation, and (ii) an innovative patch-order permutation strategy for autoregressive transformers that strengthens the role of slot vectors in reconstruction. The effectiveness of these strategies is showcased experimentally. The combined approach significantly surpasses prior slot-based autoencoder methods in unsupervised object segmentation, especially with complex real-world images. We provide the implementation code at https://github.com/gkakogeorgiou/spot . |
| title | SPOT: Self-Training with Patch-Order Permutation for Object-Centric Learning with Autoregressive Transformers |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2312.00648 |