Vision & Image

Browser interface for Stable Diffusion image generation. Prompt-to-image, inpainting, upscaling.

Face swapping and face reenactment pipeline. Swap faces in images and video with AI.

Meta's image segmentation model. Segment any object in any image with prompts.

Real-time multi-person keypoint detection. Detects body, face, hand poses from images/video.

Control diffusion models with spatial conditioning. Adds pose, edge, depth controls to image/video generation.

Programmatic video editing in Python. Cut, compose, overlay, add text/audio to videos with code.

Lip-sync videos from audio. Generates accurate lip movements matching speech audio for any video.

Text-to-video animation framework. Generates smooth animations from text prompts using diffusion models.

3D deep learning library. Renders, manipulates, and optimizes 3D meshes, point clouds, and neural radiance fields.

High-resolution 3D human reconstruction from single photo. Creates detailed textured avatars.

Expressive 3D human body model. Full body, hand, and face parametric model for avatars.