PHOTOREALISTIC HUMAN RECONSTRUCTION w/ CROSS-SCALE DIFF
Convert images to 3D models
create games with AI
Create 3D reconstructions from images using MASt3R and 3DGS
Generate a 3D model and video from images
Play a 3D breakout game
Generate textured 3D models from text descriptions
Create 3D models and videos from images
Generate 3D fractal structures using L-system rules
Create an interactive GTA-style city with collisions and NPC dialogue
Sparse-view SFM-free Gaussian Splatting in Seconds
Generate dynamic 3D torus knot shapes
Create a 3D scene with spinning lights and random torus knots
PSHuman is a cutting-edge 3D modeling tool designed for photorealistic human reconstruction. It leverages advanced AI technology, specifically cross-scale diffusion, to generate highly realistic 3D human models and videos from input images. This tool is ideal for professionals and enthusiasts alike, offering a user-friendly approach to creating lifelike 3D content.
What formats does PSHuman support for input and output?
PSHuman supports standard image formats like PNG and JPG for input. Output models are available in formats such as OBJ, FBX, and STL, with videos exported as MP4.
Can I customize the appearance of the generated 3D model?
Yes, PSHuman allows you to adjust settings like texture quality and pose to refine the output.
What are typical use cases for PSHuman?
Common applications include game development, VR avatars, film and animation, and advertising visuals.