Multi-Pig Part Detection and Association with a Fully-Convolutional Network

TLDR

Computer vision systems could enable automated, non‑invasive livestock monitoring, but the scarcity of public datasets with clear targets and evaluation metrics hampers research, leading many solutions to rely on small, private data. This study introduces a new dataset and method for instance‑level detection of multiple pigs in group‑housed environments. The approach employs a single fully‑convolutional neural network that predicts each pig’s location, orientation, body‑part positions, and pairwise associations in image space, and is trained on a new dataset of 2,000 images with 24,842 annotated pigs from 17 sites. The method attains >99% precision and >96% recall on seen environments, and 91% precision with 67% recall on unseen environments and lighting conditions, and the dataset is publicly available for download.

Abstract

Computer vision systems have the potential to provide automated, non-invasive monitoring of livestock animals, however, the lack of public datasets with well-defined targets and evaluation metrics presents a significant challenge for researchers. Consequently, existing solutions often focus on achieving task-specific objectives using relatively small, private datasets. This work introduces a new dataset and method for instance-level detection of multiple pigs in group-housed environments. The method uses a single fully-convolutional neural network to detect the location and orientation of each animal, where both body part locations and pairwise associations are represented in the image space. Accompanying this method is a new dataset containing 2000 annotated images with 24,842 individually annotated pigs from 17 different locations. The proposed method achieves over 99% precision and over 96% recall when detecting pigs in environments previously seen by the network during training. To evaluate the robustness of the trained network, it is also tested on environments and lighting conditions unseen in the training set, where it achieves 91% precision and 67% recall. The dataset is publicly available for download.