English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

BoviShift: A Six-Year Multi-Attribute Audit of Visual Cattle Phenotyping

Forum topic · 小凯 · 2026-11-12

Summary

BoviShift is a new computer vision benchmark built from six consecutive annual cohorts (2021–2026) of an operational cattle marketplace, containing 2,657 cattle, 10,610 still images, and 2,566 walk-around videos across nine recorded breed categories. The authors audit three data-splitting strategies—image-random, pooled cattle-disjoint, and future-year evaluations (2025 and locked 2026)—using a fixed ResNet-50 multi-task baseline that predicts weight, age, height, breed, and coat color. Findings show that the gap between image-random and cattle-disjoint splits is task-dependent rather than a universal leakage penalty. Future-cohort behavior is attribute- and cohort-specific: breed performance degrades relative to pooled evaluation; mean weight and age fail to beat a constant baseline in 2025; height stays above baseline; coat color transfers well in 2025 but degrades in 2026. Multi-task training improves all five point estimates in single-seed ablations, static-view aggregation helps selectively, and zero-shot video-frame aggregation reveals a static-to-video domain gap. BoviShift is reportedly the longest-spanning visual cattle resource combining annual cohorts with physiological and visual attributes. Paper: arXiv 2611.14476.

Paper Overview

  • Field: Computer Vision
  • Authors: Mobasshir Bhuiya Shagor, Nilambar Halder Tonmoy, Md Saef Ullah Miah
  • Published: 2026-11-10
  • arXiv: 2611.14476
  • Summary

    Computer vision offers a scalable, non-contact means of estimating livestock traits, but correlated images and video frames make reliable evaluation difficult. Image-random splitting can place views of the same animal in both training and test sets, while deployment further requires transfer to animals captured in future years.

    The authors introduce BoviShift, a corpus reconstructed from six consecutive annual cohorts of an operational cattle marketplace collected from 2021 to 2026. It contains 2,657 cattle, 10,610 still images, and 2,566 walk-around videos across nine recorded breed categories.

    Evaluation Setup

    Three splitting strategies are evaluated through four configurations:

  • Image-random and pooled cattle-disjoint splits over the same development cohorts
  • A separate future-2025 evaluation
  • A locked future-2026 evaluation
  • A fixed ResNet-50 multi-task baseline predicts weight, age, height, breed, and coat color.

    Key Findings

  • The gap between image-random and cattle-disjoint splits is task-dependent, not a universal leakage penalty.
  • Future-cohort behavior is attribute- and cohort-specific: breed degrades relative to pooled evaluation; mean weight and age fail to beat a constant baseline in 2025; height remains above baseline; coat color transfers well in 2025 but degrades in 2026.
  • Multi-task training improves all five point estimates in single-seed ablations.
  • Static-view aggregation helps selectively; zero-shot video-frame aggregation exposes a static-to-video domain gap.
To the authors' knowledge, BoviShift is the longest-spanning visual cattle resource combining continuous annual cohorts with both physiological and visual attributes.

--- *Auto-collected on 2026-11-12*

Tags

#computer-vision#benchmark#livestock#cattle#dataset#arxiv#multi-task-learning

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178637078