Download Gigapixel AI Download – 7.0.0

Gigapixel AI helps you do a single thing very well: increase image resolution in a natural way. This is quite difficult to do. Most upscaling algorithms don’t know how to fill in missing pixels accurately, so you end up with blocky, low-quality results.

To solve this problem, Gigapixel AI’s neural network is continuously trained on a wide variety of example images to understand what photorealistic details look like. Having learned from millions of images, it can now reliably upscale a wide variety of subjects (portraits, wildlife, landscape, architecture, etc.) and image types (DSLR, web/compressed, CG). The results are often magical.

Improve image quality

• Gigapixel AI can make your photo look like it was taken with a much higher quality camera. It has been trained on enough images of different types of subjects (such as portraits, landscapes, architecture, and more) to know how image details should look. This allows you to increase the image resolution and quality very naturally with minimal artifacts.
• The number of pixels is not the only thing that affects image quality. Normally, if you start with a low-quality image, you’ll just end up with a larger, low-quality image if you scale it up. Gigapixel AI’s technology can instead create highly believable and natural image details that can often look amazing.

Automatic face enhancement

• With our latest developments in machine learning and image recognition, we have implemented automatic face enhancement in Gigapixel AI to offer you more powerful and accurate face augmentation. You will see a switch in the right panel to enable/disable the new face enhancement feature. Face Refinement will detect very small faces (16×16 px to 64×64 px) and use targeted, enhanced upsampling through machine learning.

Image enlargement for any reason

• Gigapixel AI comes in handy for a wide range of applications. From enlarging scanned photographs, creating high-resolution crops in post-production, enlarging images to create high-quality prints, upscaling compressed images, to restoring low-resolution images from your old image libraries, Gigapixel AI succeeds where other image enlarging applications fail.

Restore true detail for actual sharpness

• Increasing the edge contrast is the easiest way to make an image look “sharp”, and how most other upscaling programs work.
• Gigapixels are different: they increase the actual sharpness of the upscaled image by restoring true detail to your image. This is only possible through training and exposure to millions of test images.

Sharp images that look like they were never upscaled

• Gigapixel AI successfully determines which image features are stars (detail) and which are noise. It then highlights and restores detail while suppressing noise in your upscaled image.

The highest quality photo enlargement you can get anywhere

• Because of these exceptional results, Gigapixel AI is used by everyone from leading photographers, to video game modders, to major motion studios.
• If you ever need to upscale your image while preserving beautiful details, take advantage of the most modern technology in the room and try Gigapixel AI.
• Upscaling hundreds of images with batch processing

A giant leap forward

• For decades, traditional photo upscaling has been achieved through algorithms, either fractal or bicubic interpolation. Interpolation algorithms work by approximating new pixels from the colors to the surrounding ones. This process cannot add any detail, resulting in blurred features. Gigapixel AI represents a huge leap forward in technology. Gigapixel AI’s neural network analyzes millions of photo pairs to determine how detail is typically lost. The network learns to create information in new images and how to enlarge, enhance and add natural details.

How to launch Gigapixel AI from Photoshop?

This plugin should be configured automatically when you install Gigapixel AI v 5.1.0 or later.
If it doesn’t, you can find our troubleshooting guide here.

Note: Gigapixel AI is only compatible with PS CC 2020 and later

What is new

There are really too many changes to get into the weeds with, so instead of an exhaustive changelog, I’ve extracted a few points.

Big changes:

• Updated AI engine to allow better compatibility and faster processing
• Added Standard and High Fidelity v2 models
• These models will eventually replace v1 models over time. If you have quality problems, please let us know.
• Updated the user interface with new design and controls
• Updated preferences with new options
• Note: This will completely reset your settings

Notable additions:

• Added pan and pinch to zoom gestures in preview
• Added updated RAW preview handling (should be closer to Photo AI)
• Added the ability to capture before/after in split and side-by-side views
• Added possibility to move the file list up and down by dragging
• Note: There is a known issue where removing/adding images causes it to revert to the default position
• Changed selection logic
• Changed controls to show mixed states if multiple images with different values ​​are selected

System Requirements:

• Windows: Windows 7, 8, 10, 11 (64-bit only)
• Mac: macOS version 10.12 (Sierra) and later
• We currently do not support Linux-based operating systems.
• Graphics drivers and OpenGL
• Your system’s graphics drivers must be fully up-to-date.
• Our apps also require OpenGL version 3.3 or later.

Hardware Requirements:

• Minimum: Intel i5 or equivalent (3.0 GHz and above) Ryzen 5 or equivalent (3.0 GHz and above)
• Recommended: Intel i7 or higher (4GHz and above) Ryzen 7 or higher (4GHz and above)
• A note about openVINO: Topaz Labs apps support Intel’s openVINO toolkit for high-speed CPU-based rendering. If you are currently using an Intel CPU, you can enable openVINO in the Preferences menu of any of our apps.

GPU/graphics card:

• Minimum: 2 GB dedicated VRAM (GT 740 or higher) 2 GB dedicated VRAM (Radeon 5870 or higher)
• Recommended: 4 GB dedicated VRAM (GTX 970 or higher) 4 GB dedicated VRAM (Radeon RX 460 or higher)
• Optimal: 8 GB dedicated VRAM (GTX 1080 or higher) 8 GB dedicated VRAM (Radeon RX 580 or higher)
• Minimum: Requirements for the application to work, users should expect slow performance, large files can cause crashes
• Recommended: Users should not experience performance issues, although slowness may occur with large files
• Optimal: Users should experience no performance issues
• * We do not support Intel HD Graphics 4600 or lower.

RAM:

• Minimum: 8 GB
• Recommended: 16GB
• Optimal: 32GB
• Monitor: A monitor that outputs with a resolution of at least 1024 x 768 is recommended.