Still image and spatial-temporal tomato data enabling detection, segmentation, tracking, and video-instance segmentation using strong and weak labels
This work provides new benchmark datasets for the agricultural computer vision community, enabling evaluation of methods for fruit detection and ripeness estimation in real-world settings.
The authors release two datasets (BUTom21 and BUTom-ST21) for tomato plant visual sensing, providing pixel-level ripeness labels and enabling detection, segmentation, tracking, and video-instance segmentation tasks. The datasets aim to support research in field-based phenotyping for horticultural crops.
In this manuscript we release two datasets for visual sensing of tomato plants grown in commercial-like settings and acquired using a robot. The first is BUTom21 which consists of still images and manual annotations. The second is BUTom-ST21 which consists of video-based data and semi-automated annotations through AI-based methods, referred to as pseudo-labels. In both cases, we provide pixel-level labels for the ripeness of the fruit. The aim is to provide the research community a challenging set of real-world imagery to explore methods to sense and estimate the state of tomato plants and their fruit, which is an important horticultural crop. Importantly, the spatial-temporal dataset provides individual fruit count and ripeness information enabling researchers to push the boundaries of field-based phenotyping.