Look Ma, No Ground Truth! Ground-Truth-Free Tuning of Structure from Motion and Visual SLAM
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913593605226496 |
|---|---|
| author | Fontan, Alejandro Civera, Javier Fischer, Tobias Milford, Michael |
| author_facet | Fontan, Alejandro Civera, Javier Fischer, Tobias Milford, Michael |
| contents | Evaluation is critical to both developing and tuning Structure from Motion (SfM) and Visual SLAM (VSLAM) systems, but is universally reliant on high-quality geometric ground truth -- a resource that is not only costly and time-intensive but, in many cases, entirely unobtainable. This dependency on ground truth restricts SfM and SLAM applications across diverse environments and limits scalability to real-world scenarios. In this work, we propose a novel ground-truth-free (GTF) evaluation methodology that eliminates the need for geometric ground truth, instead using sensitivity estimation via sampling from both original and noisy versions of input images. Our approach shows strong correlation with traditional ground-truth-based benchmarks and supports GTF hyperparameter tuning. Removing the need for ground truth opens up new opportunities to leverage a much larger number of dataset sources, and for self-supervised and online tuning, with the potential for a data-driven breakthrough analogous to what has occurred in generative AI. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_01116 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Look Ma, No Ground Truth! Ground-Truth-Free Tuning of Structure from Motion and Visual SLAM Fontan, Alejandro Civera, Javier Fischer, Tobias Milford, Michael Computer Vision and Pattern Recognition Robotics Evaluation is critical to both developing and tuning Structure from Motion (SfM) and Visual SLAM (VSLAM) systems, but is universally reliant on high-quality geometric ground truth -- a resource that is not only costly and time-intensive but, in many cases, entirely unobtainable. This dependency on ground truth restricts SfM and SLAM applications across diverse environments and limits scalability to real-world scenarios. In this work, we propose a novel ground-truth-free (GTF) evaluation methodology that eliminates the need for geometric ground truth, instead using sensitivity estimation via sampling from both original and noisy versions of input images. Our approach shows strong correlation with traditional ground-truth-based benchmarks and supports GTF hyperparameter tuning. Removing the need for ground truth opens up new opportunities to leverage a much larger number of dataset sources, and for self-supervised and online tuning, with the potential for a data-driven breakthrough analogous to what has occurred in generative AI. |
| title | Look Ma, No Ground Truth! Ground-Truth-Free Tuning of Structure from Motion and Visual SLAM |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2412.01116 |