What does this research mean for the field?

An integrated pipeline using a Light Attention Encoder-Decoder model and overhead depth imaging can accurately segment cattle bodies and estimate body weight non-invasively. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to develop a novel method for non-invasive body weight estimation in beef cattle using depth images and segmentation techniques.

June 10, 2026AnimalsOpen Access

Light Attention Encoder–Decoder for Cattle Body Segmentation and Body Weight Estimation

View Full Paper

Ask AI

Bookmark

Authors

SMSahilpreet Singh Mann HSHalah K. Shehada SASabrina T. Amorim

Discussion

Member takes

Overview

Randomized trial demonstrates non-invasive cattle body segmentation and weight estimation, suggesting improved management strategies.

Key Points

This research aims to develop a novel method for non-invasive body weight estimation in beef cattle using depth images and segmentation techniques.
Implemented a Light Attention Encoder–Decoder (LAED) model for segmentation of cattle from depth images.
Collected depth videos via an Intel RealSense D435 camera from 60 beef heifers using leave-one-animal-out cross-validation.
Used machine learning models including support vector regression and random forest for weight prediction based on extracted features.
LAED + GCT achieved a Dice coefficient of 96.91% (95% CI: 96.56–97.21%) and an IoU of 94.22% (95% CI: 93.58–94.77%).
The best weight prediction model using biometric traits and support vector regression achieved a MAPE of 6.75%, R2 of 0.68, MAE of 23.92 kg, and RMSE of 31.79 kg.
The mean signed prediction bias for the biometric-SVR model was −1.04 kg with a bootstrap 95% CI of −9.63 to 7.41 kg.

Cite This Study

Mann et al. (2026) studied this question.

synapsesocial.com/papers/6a28ffc76f82f25be989c87a https://doi.org/10.3390/ani16121773

View Full Paper

Ask AI

Bookmark

June 10, 2026AnimalsOpen Access

Light Attention Encoder–Decoder for Cattle Body Segmentation and Body Weight Estimation

View Full Paper

Ask AI

Bookmark

Authors

SMSahilpreet Singh Mann HSHalah K. Shehada SASabrina T. Amorim

Discussion

Member takes

Overview

Randomized trial demonstrates non-invasive cattle body segmentation and weight estimation, suggesting improved management strategies.

Key Points

This research aims to develop a novel method for non-invasive body weight estimation in beef cattle using depth images and segmentation techniques.
Implemented a Light Attention Encoder–Decoder (LAED) model for segmentation of cattle from depth images.
Collected depth videos via an Intel RealSense D435 camera from 60 beef heifers using leave-one-animal-out cross-validation.
Used machine learning models including support vector regression and random forest for weight prediction based on extracted features.
LAED + GCT achieved a Dice coefficient of 96.91% (95% CI: 96.56–97.21%) and an IoU of 94.22% (95% CI: 93.58–94.77%).
The best weight prediction model using biometric traits and support vector regression achieved a MAPE of 6.75%, R2 of 0.68, MAE of 23.92 kg, and RMSE of 31.79 kg.
The mean signed prediction bias for the biometric-SVR model was −1.04 kg with a bootstrap 95% CI of −9.63 to 7.41 kg.

Cite This Study

Mann et al. (2026) studied this question.

synapsesocial.com/papers/6a28ffc76f82f25be989c87a https://doi.org/10.3390/ani16121773

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark