Skip to Main content Skip to Navigation
Conference papers

Deep Active Learning from Multispectral Data Through Cross-Modality Prediction Inconsistency

Heng Zhang 1, 2 Elisa Fromont 3, 2 Sébastien Lefevre 4 Bruno Avignon 1
2 LACODAM - Large Scale Collaborative Data Mining
Inria Rennes – Bretagne Atlantique , IRISA-D7 - GESTION DES DONNÉES ET DE LA CONNAISSANCE
4 OBELIX - Observation de l’environnement par imagerie complexe
IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : Data from multiple sensors provide independent and complementary information, which may improve the robustness and reliability of scene analysis applications. While there exist many large-scale labelled benchmarks acquired by a single sensor, collecting labelled multi-sensor data is more expensive and time-consuming. In this work, we explore the construction of an accurate multispectral (here, visible & thermal cameras) scene analysis system with minimal annotation efforts via an active learning strategy based on the cross-modality prediction inconsistency. Experiments on multiple multispectral datasets and vision tasks demonstrate the effectiveness of our method. In particular, with only 10% of labelled data on KAIST multispectral pedestrian detection dataset, we obtain comparable performance as other fully supervised State-of-the-Art methods.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03236409
Contributor : Elisa Fromont <>
Submitted on : Wednesday, May 26, 2021 - 11:29:29 AM
Last modification on : Saturday, May 29, 2021 - 3:25:23 AM
Long-term archiving on: : Friday, August 27, 2021 - 7:03:14 PM

File

icip2021.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03236409, version 1

Citation

Heng Zhang, Elisa Fromont, Sébastien Lefevre, Bruno Avignon. Deep Active Learning from Multispectral Data Through Cross-Modality Prediction Inconsistency. ICIP 2021 - 28th IEEE International Conference on Image Processing, Sep 2021, Anchorage, United States. pp.1-5. ⟨hal-03236409⟩

Share

Metrics

Record views

103

Files downloads

210