Continual Learning for Image-Based Camera Localization

Shuzhe Wang, Zakaria Laskar, Iaroslav Melekhov, Xiaotian Li, Juho Kannala

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference contributionScientificvertaisarvioitu

6 Sitaatiot (Scopus)
48 Lataukset (Pure)


For several emerging technologies such as augmented reality, autonomous driving and robotics, visual localization is a critical component. Directly regressing camera pose/3D scene coordinates from the input image using deep neural networks has shown great potential. However, such methods assume a stationary data distribution with all scenes simultaneously available during training. In this paper, we approach the problem of visual localization in a continual learning setup -- whereby the model is trained on scenes in an incremental manner. Our results show that similar to the classification domain, non-stationary data induces catastrophic forgetting in deep networks for visual localization. To address this issue, a strong baseline based on storing and replaying images from a fixed buffer is proposed. Furthermore, we propose a new sampling method based on coverage score (Buff-CS) that adapts the existing sampling strategies in the buffering process to the problem of visual localization. Results demonstrate consistent improvements over standard buffering methods on two challenging datasets -- 7Scenes, 12Scenes, and also 19Scenes by combining the former scenes.
Otsikko2021 International Conference on Computer Vision, ICCV
ISBN (elektroninen)978-1-6654-2812-5
ISBN (painettu)978-1-6654-2813-2
DOI - pysyväislinkit
TilaJulkaistu - 2022
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInternational Conference on Computer Vision - Virtual, Online
Kesto: 11 lokak. 202117 lokak. 2021


NimiIEEE International Conference on Computer Vision
ISSN (painettu)1550-5499
ISSN (elektroninen)2380-7504


ConferenceInternational Conference on Computer Vision
KaupunkiVirtual, Online


Sukella tutkimusaiheisiin 'Continual Learning for Image-Based Camera Localization'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä