AI-Generated Content - Digital Human - SenseTime

z6尊龙官方网站

Apply for Trial
Core Technology
Based on the foundation of proprietary technologies and SenseCore AI infrastructure,
SenseTime has rapidly opened up AI application in multiple vertical scenarioses, and is empowering various industries.
Key Technology
  • 012D/3D Key Points Capture
  • 023D Key Points Avatar
  • 03Virtual Try-On
  • 04Digital Human
  • 05Portrait style transfer
  • 06Image/video editing

01 / 06

2D/3D Key Points Capture

2D/3D key points of face/body/gesture can be captured and tracked in various situations, e.g., darkness, occlusion, tricky angles and exaggerated expressions. The technique can be utilized for rendering special effects in the post processing.

02 / 06

3D Key Points Avatar

By detecting the 3D key points of face/body/hand, the SDK can drive the expressions/body movements/gestures for 2D/3D avatar. Additionally, it works well for multiple expressions and complex movements.

03 / 06

Virtual Try-On

2D/3D key point detection and tracking for body (hands/feet) and post-rendering, it can realize the effect of virtual manicure and virtual shoe fitting. Furthermore, it has a robust visualization under complex situations such as different light intensities and occlusion.

04 / 06

Digital Human

Fast customized generation of animatable photorealistic 2D real speaker and 3D virtual character assets, and vivid animation driven by audio with deep neural network. Also support human machine interaction with advanced mult-modal algorithms.

05 / 06

Portrait style transfer

Style transfer can convert the face in real-time video or high-resolution pictures to a specific style of portrait, including Japanese manga style, CG style, hand-painted style, etc.

06 / 06

Image/video editing

Based on image generation technologies such as generative adversarial network, it can edit pictures/videos in detail, and support downstream tasks such as makeup editing, hair editing, motion transfer, etc.

Technology
  • 2D/3D Key Points Capture
  • 3D Key Points Avatar
  • Virtual Try-On
  • Digital Human
  • Portrait style transfer
  • Image/video editing

01 / 06

2D/3D Key Points Capture

2D/3D key points of face/body/gesture can be captured and tracked in various situations, e.g., darkness, occlusion, tricky angles and exaggerated expressions. The technique can be utilized for rendering special effects in the post processing.

02 / 06

3D Key Points Avatar

By detecting the 3D key points of face/body/hand, the SDK can drive the expressions/body movements/gestures for 2D/3D avatar. Additionally, it works well for multiple expressions and complex movements.

03 / 06

Virtual Try-On

2D/3D key point detection and tracking for body (hands/feet) and post-rendering, it can realize the effect of virtual manicure and virtual shoe fitting. Furthermore, it has a robust visualization under complex situations such as different light intensities and occlusion.

04 / 06

Digital Human

Fast customized generation of animatable photorealistic 2D real speaker and 3D virtual character assets, and vivid animation driven by audio with deep neural network. Also support human machine interaction with advanced mult-modal algorithms.

05 / 06

Portrait style transfer

Style transfer can convert the face in real-time video or high-resolution pictures to a specific style of portrait, including Japanese manga style, CG style, hand-painted style, etc.

06 / 06

Image/video editing

Based on image generation technologies such as generative adversarial network, it can edit pictures/videos in detail, and support downstream tasks such as makeup editing, hair editing, motion transfer, etc.

Business
Partnership

Apply for Trial

Fill out this short form, and we’ll get back to you as soon as possible.

Apply for Trial

Business Partnership

Email: business@qidebank.com