Tailored Data Service for Retail

Nexdata supports object image and 2D/3D face anti-spoofing data collection as well as cross-camera tracking data collection by simulating real-world scenarios. We have extensive data service experience in face recognition, object detection, semantic segmentation, and target behavior recognition.

Face Recognition

Support 2D&3D face recognition and anti-spoofing data collection. Provide data labeling of bounding box, keypoints, segmentation etc.

Human Behavior Recognition

Support multi-device, multi-scene, multi-angle human behavior data collection/labeling for human locations, behaviors and attributes.

Store Traffic Counting

Support data collection of people flow data in a wide range of monitoring scenarios and data annotation of people distribution.

Search Relevance

Support data labeling of bounding boxes, segmentation, classification for objects in image.

Virtual Try-on

Support data collection of clothing and accessories, and provide data labeling of classification, segmentation and attribute.

Inventory Management

Supports data collection of warehouse and logistics environment, and provide data collection of classification, segmentation and attribute.

Compliance & Security

Nexdata place the utmost emphasis on data security and client trust. We follows Personal Information Protection Act, GDPR, CCPA, PIPC and HIPAA regulations. we have also achieved ISO 27001,ISO 27701 and ISO 9001 qualifications for security and regulatory compliance. Nexdata delivers unparalleled data security, earning the trust of our clients through our adherence to these globally recognized standards.

Case Studies

USE CASE：Facial Recognition Payment.
CHALLENGE：Clients need to improve the accuracy of face recognition on the payment platform and have high requirements on face data type and environmental lighting.
SOLUTION：Nexdata collected 2D/3D face data and face anti-spoofing data under different lighting, angles, occlusion, and perfromed keypoint annotation and facial feature annotation for collected data.

Automatic Identification of Unmanned Vending Machine

USE CASE：Automatic Identification of Unmanned Vending Machine.
CHALLENGE：Client wants to further enhance the automatic identification of the merchandise in the unmanned containers.
SOLUTION：Nexdata simulates the real scene by installing a camera in the unmanned container and collects 300,000 pictures. Due to the high difficulty of manual judgement on a wide range of goods, the labeling efficiency and accuracy rate were greatly improved by searching for labels and displaying multi-angle picture samples for each category of labels.

USE CASE：Virtual Clothes Fitting System.
CHALLENGE：Clients hopes to improve the recognition accuracy of the smart fitting system and has very high requirements for the accuracy of fashion segmentation. It needs to segment and label a large amount of character and clothing data and add attributes.
SOLUTION：Nexdata's refined matting template, with polygon and object fit within 3px, completely meets the customer's requirements for annotation precision and accuracy.

USE CASE：E-commerce OCR System.
CHALLENGE：The customer needed an OCR model to identify key information on product images. There are 5 languages in the pictures that need to be annotated, and there are blur, large-area occlusion, tilted text, etc. The annotation accuracy rate is required to reach 95%.
SOLUTION：Nexdata completed the data annotation of 15,000 OCR images within the specified time, with an accuracy rate of 98%.

Recommended Dataset

4,001 People Single Object Multi-view Tracking Data

4,001 People Single Object Multi-view Tracking Data, the data collection site includes indoor and outdoor scenes (such as supermarket, mall and community, etc.) , where each subject appeared in at least 7 cameras. The data diversity includes different ages, different time periods, different cameras, different human body orientations and postures, different collecting scenes. It can be used for computer vision tasks such as object detection and object tracking in multi-view scenes.

Multi-view tracking Surveillance scenes Human body Multiple scenes

212 People – 48,000 Images of Multi-person and Multi-view Tracking Data

212 People – 48,000 Images of Multi-person and Multi-view Tracking Data. The data includes males and females, and the age distribution is from children to the elderly. The data diversity includes different age groups, different shooting angles, different human body orientations and postures. For annotation, we adpoted rectangular bounding boxes annotations on human body. This dataset can be used for multiple object tracking and other tasks.

multiple object tracking surveillance scene human body detection security data smart security data

10,114 People Multi-view Tracking Data

This data is a multi-view tracking data of 10,114 people in surveillance scenes. Surveillance scenes includes indoor and outdoor scenes. The data includes men and women of different ages. In terms of annotation, the human body bounding boxes, human body + riding object bounding boxes, and 21 human body attributes of tracking objects were annotated. This data can be used for human body multi-view tracking, Re-ID and other tasks.

Multi-view Cross-camera human body tracking Re-ID Re-ID

5,521 People - Re-ID Data in Surveillance Scenes

5,521 People - Re-ID Data in Surveillance Scenes. The data includes indoor scenes and outdoor scenes. The data includes males and females, and the age distribution is from children to the elderly. The data diversity includes different age groups, different time periods, different shooting angles, different human body orientations and postures, clothing for different seasons. For annotation, the rectangular bounding boxes and 15 attributes of human body were annotated. The data can be used for re-id and other tasks.

Surveillance scenes Re-ID different age groups different time periods different shooting angles different human body orientations and postures clothing for different seasons human body rectangular bounding boxes attributes annotation

10,034 People - Re-ID Data in Surveillance Scenes

10,034 People - Re-ID Data in Surveillance Scenes. The data includes supermarket (inside supermarket and at the gate of the supermarket) scenes. The data includes males and females and the age distribution is from children to the elderly. In this dataset, the rectangular bounding boxes and 15 attributes of human body were annotated.The data can be used for re-id and other tasks.

Surveillance scenes Re-ID multiple time periods multiple ages human body rectangular bounding boxes attributes annotation

11,130 People - Re-ID Data in Real Surveillance Scenes

11,130 People - Re-ID Data in Real Surveillance Scenes. The data includes indoor scenes and outdoor scenes. The data includes males and females, and the age distribution is from children to the elderly. The data diversity includes different age groups, different time periods, different shooting angles, different human body orientations and postures, clothing for different seasons. For annotation, the rectangular bounding boxes and 15 attributes of human body were annotated. This data can be used for re-id and other tasks.

different ages different time periods different cameras different human body orientations and postures different ages collecting environment

2,769 People - CCTV Re-ID Data in Europe

2,769 People – CCTV Re-ID Data in Europe. The data includes males and females, the race distribution is Caucasian, black, Asian, and the age distribution is from children to the elderly. The data diversity includes different age groups, different time periods, different cameras, different human body orientations and postures. For annotation, the rectangular bounding boxes and 15 attributes of human body were annotated. This data can be used for re-id and other tasks.

European CCTV Re-ID different age groups different time periods different cameras different human body orientations and postures the human body rectangular bounding boxes the rectangular bounding boxes

600,000 Images – Vehicle Re-ID Data in Surveillance Scenes

600,000 Images – Vehicle Re-ID Data in Surveillance Scenes. The collecting scenes of this dataset include outdoor roads (highways, road bayonets, urban roads, etc.). The data diversity includes different cameras, multiple outdoor scenes, multiple time periods. For annotation, rectangular bounding boxes of vehicles were annotated. The data can be used for tasks such as vehicle re-id in surveillance scenes.

Vehicle Re-ID outdoor roads different cameras multiple outdoor scenes multiple time periods rectangular bounding boxes of vehicles

Why Nexdata

One-stop Data Service

Nexdata.ai provides comprehensive data
annotation and collection services to help
you succeed with your AI projects.

Data QA System

Nexdata delivers high-quality data with
intelligent self-inspection, multiple quality
checks, and ISO9001 certification.

Rich Annotation Tools

30 proven annotation tools for full coverage of
voice, image video, 3D point cloud and text
data annotation requirements.

Compliance & Security

We follows Personal Information Protection
Act, GDPR, ISO27001/ISO27701 for security
and regulatory compliance.

AI-assisted Pre-recognition

With the help of AI-assisted pre-recognition
function, human-computer interaction
semi-automatic annotation is realized.

Retail Data
Solutions