Back to datasets
Dataset assetOpen Source CommunityComputer VisionAnimal Behavior Recognition

stockeh/dog-pose-cv

The dataset contains 20,578 images of dogs in various poses, labeled as ‘standing’, ‘sitting’, ‘lying down’, or ‘undefined’. It is intended for computer‑vision tasks that identify dog behavior from images. The images span 120 dog breeds with varying resolutions; 50 % of the images have resolutions between 361 × 333 and 500 × 453 pixels. The dataset is adapted from the Stanford Dog Dataset with re‑labeled poses. Class distribution is imbalanced, with ‘lying down’ nearly double the ‘sitting’ images, and ‘undefined’ mainly consisting of close‑up portraits, which may limit processing of such images. Users should consider class‑balancing techniques such as oversampling or data augmentation.

Source
hugging_face
Created
Nov 28, 2025
Updated
Mar 30, 2024
Signals
195 views
Availability
Linked source ready
Overview

Dataset description and usage context

Dataset Overview

Basic Information

  • License: Apache‑2.0
  • Task Category: Image Classification
  • Language: English
  • Dataset Size: 10K < n < 100K
  • Dataset Name: DogPoseCV

Dataset Content

  • Number of Images: 20,578
  • Image Description: Dogs in various poses, labeled standing, sitting, lying down, undefined
  • Number of Breeds: 120
  • Image Resolution: 50 % of images have resolutions between 361 × 333 and 500 × 453 pixels

Dataset Structure

  • Class Distribution:
    • standing: 4,143 images
    • sitting: 3,038 images
    • lying down: 7,090 images
    • undefined: 6,307 images

Data Collection & Processing

  • Source: Adapted from the Stanford Dog Dataset
  • Labels: Manually annotated as standing, sitting, lying down, undefined

Bias, Risks, and Limitations

  • Class Imbalance: Noticeable imbalance, with "lying down" nearly twice the "sitting" count.
  • Limitation: The "undefined" category mainly contains hard‑to‑distinguish pose images such as close‑up portraits.

Usage Recommendations

  • Handling Advice: Address class imbalance using oversampling or data‑augmentation techniques during model training.
Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio