Today's Image & Video Generation: Fastest-Growing Projects — July 23, 2026
Today's the Image & Video Generation space on GitHub, there's a notable trend towards AI-driven enhancements and automation for both photographers and designers. Tools ranging from intelligent image enhancement extensions to advanced video temporal grounding systems are gaining traction among developers and professionals alike. One standout project is Kirakun0328/text-to-vrma, which leverages text input to generate VRM-compatible motion data, a specialized application in the virtual reality space.
Kirakun0328's text-to-vrma translates textual descriptions into VRM-specific motion data for use in virtual environments. With 90 stars and a growth score of 37.83, it’s clear that developers are interested in this innovative approach to creating realistic character animations from text inputs.
FuelMagistrateLead's photoshop-ai-smart-enhance is an AI-powered extension designed to enhance images within Adobe Photoshop, offering features such as exposure correction and noise reduction through machine learning algorithms. This project has seen significant growth with 109 stars and a growth score of 30.36, likely due to its practical application in the professional photography and design community.
Finderchangchang's codex-autoskin is an agent-native skin engine that enables users to automatically apply new skins or designs to their Codex desktop app using just one image input. The project has garnered 107 stars and a growth score of 20.93, highlighting its potential for streamlining design processes through automation.
Nexu-io's codex-slides is an open-source AI slide studio integrated into the Codex platform, offering rapid and visually rich presentation creation with support for real-time previewing and editing. With 107 stars and a growth score of 18.65, this tool is growing due to its innovative approach to visual content creation and ease of use.
Raghurajeshc134's Sapphire-PS-AI-Compositor introduces AI-powered tools for compositing images in Photoshop, enhancing workflow efficiency for professional photographers and designers. This project has 152 stars and a growth score of 15.04, indicating its value in the market for advanced image editing solutions.
Abnormal-codex's Ai-Pixel-Design-Archive is described as an AI-driven image generator tailored specifically for professional design purposes, aiming to streamline the creation process with advanced generation capabilities. With 151 stars and a growth score of 15.02, this tool is gaining traction among designers looking for sophisticated image generation solutions.
MCG-NJU's TimeLens2 focuses on video temporal grounding using multimodal large language models (LLMs), facilitating better understanding and interaction with video content through AI. With 39 stars and a growth score of 12.86, it is growing due to its innovative approach in handling video data with advanced linguistic capabilities.
OpenSenseNova's SenseNova-Vision is an ambitious project aiming to unify vision tasks under a single multimodal generation framework. It has attracted significant attention with 465 stars and a growth score of 11.81, indicating strong interest from the developer community for its broad application potential in image processing.
Cdfan0627's LongE2V is an advanced system for long-horizon event-based video reconstruction, prediction, and frame interpolation using video diffusion models, as showcased at SIGGRAPH 2026. With a growth score of 2.50 and 34 stars, it appears to be in the early stages of adoption but demonstrates promising research potential.
These projects collectively showcase the dynamic nature of AI-driven image and video generation tools, each addressing unique aspects of visual content creation and enhancement through innovative use of machine learning techniques.
Kirakun0328's text-to-vrma translates textual descriptions into VRM-specific motion data for use in virtual environments. With 90 stars and a growth score of 37.83, it’s clear that developers are interested in this innovative approach to creating realistic character animations from text inputs.
FuelMagistrateLead's photoshop-ai-smart-enhance is an AI-powered extension designed to enhance images within Adobe Photoshop, offering features such as exposure correction and noise reduction through machine learning algorithms. This project has seen significant growth with 109 stars and a growth score of 30.36, likely due to its practical application in the professional photography and design community.
Finderchangchang's codex-autoskin is an agent-native skin engine that enables users to automatically apply new skins or designs to their Codex desktop app using just one image input. The project has garnered 107 stars and a growth score of 20.93, highlighting its potential for streamlining design processes through automation.
Nexu-io's codex-slides is an open-source AI slide studio integrated into the Codex platform, offering rapid and visually rich presentation creation with support for real-time previewing and editing. With 107 stars and a growth score of 18.65, this tool is growing due to its innovative approach to visual content creation and ease of use.
Raghurajeshc134's Sapphire-PS-AI-Compositor introduces AI-powered tools for compositing images in Photoshop, enhancing workflow efficiency for professional photographers and designers. This project has 152 stars and a growth score of 15.04, indicating its value in the market for advanced image editing solutions.
Abnormal-codex's Ai-Pixel-Design-Archive is described as an AI-driven image generator tailored specifically for professional design purposes, aiming to streamline the creation process with advanced generation capabilities. With 151 stars and a growth score of 15.02, this tool is gaining traction among designers looking for sophisticated image generation solutions.
MCG-NJU's TimeLens2 focuses on video temporal grounding using multimodal large language models (LLMs), facilitating better understanding and interaction with video content through AI. With 39 stars and a growth score of 12.86, it is growing due to its innovative approach in handling video data with advanced linguistic capabilities.
OpenSenseNova's SenseNova-Vision is an ambitious project aiming to unify vision tasks under a single multimodal generation framework. It has attracted significant attention with 465 stars and a growth score of 11.81, indicating strong interest from the developer community for its broad application potential in image processing.
Cdfan0627's LongE2V is an advanced system for long-horizon event-based video reconstruction, prediction, and frame interpolation using video diffusion models, as showcased at SIGGRAPH 2026. With a growth score of 2.50 and 34 stars, it appears to be in the early stages of adoption but demonstrates promising research potential.
These projects collectively showcase the dynamic nature of AI-driven image and video generation tools, each addressing unique aspects of visual content creation and enhancement through innovative use of machine learning techniques.