Finding Waldo: Towards Efficient Exploration of NeRF Scene Spaces
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917608147648512 |
|---|---|
| author | Skartados, Evangelos Yucel, Mehmet Kerim Manganelli, Bruno Drosou, Anastasios Saà-Garriga, Albert |
| author_facet | Skartados, Evangelos Yucel, Mehmet Kerim Manganelli, Bruno Drosou, Anastasios Saà-Garriga, Albert |
| contents | Neural Radiance Fields (NeRF) have quickly become the primary approach for 3D reconstruction and novel view synthesis in recent years due to their remarkable performance. Despite the huge interest in NeRF methods, a practical use case of NeRFs has largely been ignored; the exploration of the scene space modelled by a NeRF. In this paper, for the first time in the literature, we propose and formally define the scene exploration framework as the efficient discovery of NeRF model inputs (i.e. coordinates and viewing angles), using which one can render novel views that adhere to user-selected criteria. To remedy the lack of approaches addressing scene exploration, we first propose two baseline methods called Guided-Random Search (GRS) and Pose Interpolation-based Search (PIBS). We then cast scene exploration as an optimization problem, and propose the criteria-agnostic Evolution-Guided Pose Search (EGPS) for efficient exploration. We test all three approaches with various criteria (e.g. saliency maximization, image quality maximization, photo-composition quality improvement) and show that our EGPS performs more favourably than other baselines. We finally highlight key points and limitations, and outline directions for future research in scene exploration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_04508 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Finding Waldo: Towards Efficient Exploration of NeRF Scene Spaces Skartados, Evangelos Yucel, Mehmet Kerim Manganelli, Bruno Drosou, Anastasios Saà-Garriga, Albert Computer Vision and Pattern Recognition Graphics Neural Radiance Fields (NeRF) have quickly become the primary approach for 3D reconstruction and novel view synthesis in recent years due to their remarkable performance. Despite the huge interest in NeRF methods, a practical use case of NeRFs has largely been ignored; the exploration of the scene space modelled by a NeRF. In this paper, for the first time in the literature, we propose and formally define the scene exploration framework as the efficient discovery of NeRF model inputs (i.e. coordinates and viewing angles), using which one can render novel views that adhere to user-selected criteria. To remedy the lack of approaches addressing scene exploration, we first propose two baseline methods called Guided-Random Search (GRS) and Pose Interpolation-based Search (PIBS). We then cast scene exploration as an optimization problem, and propose the criteria-agnostic Evolution-Guided Pose Search (EGPS) for efficient exploration. We test all three approaches with various criteria (e.g. saliency maximization, image quality maximization, photo-composition quality improvement) and show that our EGPS performs more favourably than other baselines. We finally highlight key points and limitations, and outline directions for future research in scene exploration. |
| title | Finding Waldo: Towards Efficient Exploration of NeRF Scene Spaces |
| topic | Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2403.04508 |