Uncertainty Driven Active Learning for Image Segmentation in Underwater Inspection
This work addresses the high cost of data labeling for underwater inspection tasks, but it is incremental as it applies existing active learning methods to a specific domain.
The paper tackles the problem of reducing data labeling costs for image segmentation in underwater infrastructure inspection by applying active learning with mutual information acquisition via Monte Carlo dropout. It shows that using active learning with HyperSeg on a pipeline inspection dataset of over 50,000 images achieves 67.5% meanIoU with 12.5% of the data, compared to 61.4% with random selection.
Active learning aims to select the minimum amount of data to train a model that performs similarly to a model trained with the entire dataset. We study the potential of active learning for image segmentation in underwater infrastructure inspection tasks, where large amounts of data are typically collected. The pipeline inspection images are usually semantically repetitive but with great variations in quality. We use mutual information as the acquisition function, calculated using Monte Carlo dropout. To assess the effectiveness of the framework, DenseNet and HyperSeg are trained with the CamVid dataset using active learning. In addition, HyperSeg is trained with a pipeline inspection dataset of over 50,000 images. For the pipeline dataset, HyperSeg with active learning achieved 67.5% meanIoU using 12.5% of the data, and 61.4% with the same amount of randomly selected images. This shows that using active learning for segmentation models in underwater inspection tasks can lower the cost significantly.