Skip to content

Releases: Kiri-Innovation/3D-Scan-Preparation-Tool-by-KIRI-Engine

3D Scan Prep Tool v1.0.1

24 Jun 10:12
@bluenile3d bluenile3d
v1.0.1
ebad124

Choose a tag to compare

View all tags
3D Scan Prep Tool v1.0.1 Latest
Latest

3D Scan Prep Tool v1.0.1

Bugfix release.

  • Fixed 360 video extraction when using Every Nth frame.
  • Sparse 360 extraction now uses fast frame seeking instead of scanning the full source video.
  • This fixes jobs appearing stuck at “Extracting video frames... decoding source video 0/2”.
  • Better UI scaling
Assets 6
Loading

3D Scan Prep Tool v1.0.0

11 Jun 04:00
@bluenile3d bluenile3d
v1.0.0
62df693
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Choose a tag to compare

View all tags
3D Scan Prep Tool v1.0.0

Initial public release of 3D Scan Prep Tool by KIRI Engine.

3D Scan Prep Tool is an open-source desktop app for preparing images, videos, 360 footage, masks, and scan-focused processing for photogrammetry and 3D scanning workflows.

Downloads

Windows

Works on 64-bit Windows 10/11.

The Windows CUDA build is split into multiple installer files because the full GPU-enabled app is large.

Download all Windows installer parts into the same folder:

  • 3D Scan Prep Tool.exe
  • 3D Scan Prep Tool-1.bin
  • 3D Scan Prep Tool-2.bin
  • any other matching .bin parts, if present

Then run 3D Scan Prep Tool.exe

Do not run or move the .bin files separately. They must stay beside the installer .exe.

The _AI_Models-v1.0.0.zip file is only needed if you are building from source.

The installers include the AI models.

For source builds, extract it into the repository root so the folder is:

_AI_Models/

Features

Image folder preparation for photogrammetry datasets
Blur/sharpness isolation tools
Video frame extraction
360/equirectangular video view extraction
AI mask generation
Mask preview
RAW and HEIC input support
Scan-focused contrast adjustment
Local contrast boost
Exposure-fusion look
Feature sharpening
NVIDIA CUDA acceleration on supported Windows systems
CPU fallback for unsupported hardware
Hardware Notes
The Windows build uses PyTorch CUDA 12.8.

Expected Windows behavior:

NVIDIA RTX 20 / 30 / 40 / 50 class GPUs should use CUDA acceleration when drivers support it.
RTX 50 / Blackwell support requires this CUDA 12.8 build.
Older NVIDIA GTX 10-series / Pascal GPUs will likely fall back to CPU.
AMD GPUs, Intel GPUs, unsupported NVIDIA GPUs, and systems without a compatible GPU should fall back to CPU.
CPU fallback is slower, especially for AI masking, but should keep the app usable.

Safety Note

Always keep a backup of important source images and videos before running batch processing.
Some workflows can move source images into sorting or review folders when enabled.

Loading