Log - Photoscan to Game-Ready

This project was NOT created with AI tools

This project shows how I create optimized, game-ready meshes from photoscans with high-quality images shot in nature.

I found this log in a forest and thought it would be ideal for a photoscan, so I grabbed my phone camera and shot over 50 images of it from all different angles. At this point you don't know if you made a good job with catching all the detail on your photos. At home, I used a photogrammetry software to create a mesh from the images und went through the shown pipeline until I eventually ended up with a game-ready mesh, usable for environments in game engine with LODs and greypacked textures.
Enjoy!

Step 1: RAW images from nature

Step 1: RAW images from nature

Step 2: Sparse cloud, point cloud, mesh, textured mesh generation

Step 2: Sparse cloud, point cloud, mesh, textured mesh generation

Step 3: Clean-Up of high-poly in Blender

Step 3: Clean-Up of high-poly in Blender

Step 4: Creation of low-poly mesh and clean unwrap

Step 4: Creation of low-poly mesh and clean unwrap

Step 5: Baking high-poly color details

Step 5: Baking high-poly color details

Step 6: Flattening color map - remove shadows, remove color artifacts

Step 6: Flattening color map - remove shadows, remove color artifacts

Step 7: Bending ground plane for better use in Engine

Step 7: Bending ground plane for better use in Engine

Step 8: Roughness, AO, Normal creation in Substance 3D Painter

Step 8: Roughness, AO, Normal creation in Substance 3D Painter

Step 9: LOD creation, Greypacking textures - for performance

Step 9: LOD creation, Greypacking textures - for performance

LOD 0 geometry and unwrap

LOD 0 geometry and unwrap