Popular products
speech recognition
Face recognition
Character recognition
Content security
Analysis of emotional tendency
information community file
Console
Portrait segmentation
Identify the human body contour in the image, separate it from the background, and adapt to single person, multiple human bodies, complex background, and various human postures; Widely used in portrait matting and beautification, photo background replacement, ID photo production, privacy protection and other scenes
Juhui
Complete enterprise certification, online API can enjoy up to 5 QPS, 50000 times Free test call
Use Now
Technical Documentation
product price
common problem
Privatization deployment scheme
Function introduction
Function demonstration
Application scenarios
Product advantages
product price
Usage
Instant free experience of portrait segmentation
Related recommendations
Function introduction
Universal Portrait Segmentation
The human body contour is separated from the image background, and the segmented binary image, gray-scale image, and transparent background portrait foreground image are returned. It supports multiple human bodies, complex backgrounds, occlusion, back, side, and other human postures
Function demonstration
Application scenarios
Portrait matting and beautification
Human special effects
Post processing of film and television
Portrait matting and beautification
The portrait in the original picture is separated, and a new background image is selected for replacement and synthesis; At the same time, the background can be virtualized to highlight the portrait and achieve a large aperture portrait photography effect
Cooperation cases
Mira camera
Product advantages
Leading algorithm
Portrait segmentation algorithm is the industry leader, evaluating more than 90% of IoU, and continuous iterative optimization based on application feedback and scene data
Flexible application
It supports the return of segmented binary images, grayscale images, and portrait foreground images, and can independently set which result images to return through parameter settings to avoid wasting bandwidth
Stable service
It can provide enterprise level stable and accurate mass traffic service, with millisecond level identification and response capability and 99.9% reliability guarantee
product price
Free version
Applicable to individual developers and enterprises during the test period
Free test amount
Up to 50000 times
Concurrency support
Maximum 5QPS
Customer service response
Within 5 working days
Price
zero
element
Free use
Paid version
It is applicable to the official landing of enterprise users
Paid transfer amount
unlimited
QPS configuration
Flexible purchase
Customer service response
7 * 24 hours
Monthly rent
200-500
RMB/QPS
Subscription payment
Custom Edition
Applicable to key customers who need special cooperation mode
Enjoy special price for key customers
Purchase a large number of QPS
Other payment mode purchase
Business consulting
Pricing Instructions
The online API for portrait segmentation is billed by QPS (concurrent requests per second). QPS billing adopts segmented ladder pricing. QPS can be purchased and expanded at any time according to business needs. QPS can be purchased monthly or separately by day
Price Document
Usage
Online interface
Online portrait segmentation interface is provided to quickly identify the contour range of the human body, separate it from the background, and go to the console for free use
Use Now
API documentation
Privatization deployment
Deployed to a local server or a proprietary cloud server (GPU is required) to realize the portrait segmentation function, support distributed high concurrency business processing, and apply to such scenarios as the composition of government documents
Scheme introduction
Apply for free trial
Instant free experience of portrait segmentation
Public cloud API can enjoy up to 5 QPS and 50000 free test calls
Use Now
Related recommendations
Zero threshold AI development self-service platform
Self help training can complete AI development with zero foundation, covering images, text, voice, video and other scenes, and can be completed in 15 minutes at the earliest
Human body detection and attribute recognition
Detect all human bodies in the image, and identify more than 20 types of attribute information, including gender, age, clothing wearing, behavior, etc
People flow statistics
Count the number of human bodies and flow trend in the image, and take the head and shoulder as the target to identify the number of people, without face and full body photos
gesture recognition
Recognize 24 common gestures, such as OK, fist, heart comparison, like, and number, and support self taking, others taking, and various shooting angles
Hand key point recognition
Locate 21 main bone nodes of the hand, which can be used to customize gesture detection, AR special effects, human-computer interaction and other scenes
Driving behavior analysis
Identify drivers' violations such as using mobile phones, smoking, not wearing seat belts, and leaving the steering wheel with both hands according to the on-board scene
Technical capability
Voice technology
Character recognition
Face and Human Body
Image technology
Language and knowledge
video technique
View all AI capabilities
Scenario application
Cooperation consulting
Pre sales consultation
Fill in your business needs, and the exclusive account manager will contact you as soon as possible to provide one-on-one consulting services
After sales intelligent assistant
Intelligent diagnosis to quickly solve the use problem
Contact Sales
For more information, please call 400-920-8999 to 1
Experience AI
Web end to AI Competency Experience Center
The mobile terminal opens Baidu APP "Scan"