T-shirt Occlusion Image Dataset
#object detection #image classification #image recognition #product detection #computer vision
- 5000 records
- 1.2G
- JPG/PNG/JSON
- CATL
MOBIUSI INC
Updated:2026-04-18
AI Analysis & Value Prop
The T-shirt Occlusion Image Dataset addresses the challenges faced in the retail e-commerce industry, particularly in recognizing clothing items that are partially obstructed or affected by varying lighting conditions. Existing solutions often struggle with accurately identifying products in complex images where occlusion is a factor, leading to inefficiencies in automated product recognition systems. This dataset aims to provide a robust collection of images to train models that can effectively handle occlusions and improve product visibility in online shopping environments. Data collection was carried out using a high-resolution camera in diverse lighting scenarios with multiple subjects wearing various T-shirts. Quality control measures included multiple rounds of annotation, consistency checks among annotators, and expert reviews to ensure high accuracy. The images are stored in JPG format, organized by categories of occlusion and lighting conditions.
Dataset Insights
Sample Examples
7ecbcaa5**.jpg|2080*961|128.93 KB
b66824a0**.jpg|1224*1632|234.31 KB
653c7ab8**.jpg|2080*961|109.79 KB
8bfb7569**.jpg|1224*1632|350.95 KB
7836ab87**.jpg|1224*1632|357.26 KB
8bcf3850**.jpg|1632*1224|98.30 KB
48b73fc8**.jpg|1224*1632|257.37 KB
39224463**.jpg|1186*1685|82.61 KB
aa094efc**.jpg|1224*1632|142.77 KB
d94e6c22**.jpg|976*1200|555.15 KB
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| quality | string | Resolution |
| image_brightness | float | Used to assess the overall brightness level of an image. |
| image_contrast | float | Used to evaluate the contrast level of an image. |
| image_sharpness | float | Used to assess the clarity of an image. |
| occlusion_level | float | Describes the proportion of the T-shirt that is obscured in the image. |
| shadow_presence | boolean | Determines whether shadows are present in the image. |
| wrinkle_presence | boolean | Checks if there are noticeable wrinkles on the T-shirt. |
| tshirt_color | string | Indicates the main color of the T-shirt visible in the image. |
| background_clutter_level | float | Measures the level of clutter in the image's background. |
| lighting_condition | string | Refers to the lighting conditions during image capture (e.g., natural light, indoor light, etc.). |
| angle_of_view | string | Refers to the viewpoint from which the T-shirt image is captured (e.g., front, side, etc.). |
| image_saturation | float | Used to evaluate the saturation of colors in an image. |
Compliance Statement
| Authorization Type | Proprietary - Commercial AI Training License (No Redistribution) |
| Commercial Use | Requires exclusive subscription or authorization contract (monthly or per-invocation charging) |
| Privacy and Anonymization | No PII, no real company names, simulated scenarios follow industry standards |
| Compliance System | Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs |
Frequently Asked Questions
- What is the T-shirt Occlusion Recognition Image Dataset?
- The T-shirt Occlusion Recognition Image Dataset is a collection of images designed for object detection tasks to improve accuracy in retail e-commerce by addressing issues caused by product occlusion and lighting interference.
- How does this dataset help improve product recognition accuracy in e-commerce?
- By providing images of T-shirts under various occlusion and lighting conditions, this dataset aids in training more robust machine learning models, thereby improving product recognition accuracy.
- What types of images are included in this dataset?
- The dataset includes images of T-shirts that may have occlusions from other objects or variations in lighting.
- Why does the retail industry need such a dataset?
- In the retail industry, enhancing product recognition system accuracy is crucial for reducing misjudgments and improving user experience, and such datasets help models learn to handle complex real-world conditions.
- What type of machine learning tasks is this dataset suitable for?
- This dataset is suitable for object detection tasks, particularly those requiring identification and localization of products affected by occlusion.
Can't find the data you need?
Post a request and let data providers reach out to you.
Get this Dataset
Verified for Enterprise UseCite this Work
@dataset{Mobiusi2025,
title={T-shirt Occlusion Image Dataset},
author={MOBIUSI INC},
year={2025},
url={https://www.mobiusi.com/datasets/e7e45f0d0e89240b2a0c4d7c8ea87cf1},
urldate={2025-08-28},
keywords={T-shirt occlusion dataset,image recognition dataset,retail e-commerce data,computer vision dataset},
version={1.0}
}Using this in research? Please cite us.
placeholder
placeholder
placeholder
placeholder
placeholder
placeholder
placeholder