Today's Image & Video Generation: Fastest-Growing Projects — October 28, 2026
Today's the Image & Video Generation space, there's a noticeable trend towards specialized applications that cater to niche markets and creative industries. The development of AI-driven tools continues to evolve rapidly, with developers focusing on unique use cases such as adult content generation, manga creation, and video storyboard visualization. These projects are gaining traction due to their innovative approaches and practical implementations.
The tool dsh-video-storyboard by SAOS-3624 is designed for generating scene-aware storyboards in a single image format that includes timestamps, transcripts, and audio envelopes. This repository has seen significant growth with a score of 78.00, likely due to its comprehensive approach to visualizing video content in a concise manner.
adult-ai-video by wlnhnoqefoxt is an AI-driven generator for creating adult-themed videos from text descriptions and images. With a high growth score of 60.80 and over 200 stars, this project has attracted considerable attention, possibly due to its ability to generate NSFW content at a high resolution.
manga-generator by stockedge is a tool designed for manga creation, handling everything from scriptwriting to image generation and PDF output. It has gained moderate interest with 119 stars, indicating a strong community engagement among manga enthusiasts looking for AI-driven solutions to streamline the creative process.
pixel-loom by yangsir2077 is an open-source pixel art and animation editor that supports MCP-based AI creation of visual content. Despite having fewer stars at 89, its high commit activity (31 commits in the last month) suggests active development and user engagement within a dedicated community.
iris-3b by speridlabs is a diffusion transformer capable of direct pixel-space generation without VAE, along with fine-tuned capabilities for monocular depth estimation and image restoration. With 206 stars and a growth score of 7.35, this project stands out due to its versatile applications in computer vision tasks.
moss-image-forge-neo by eduardoabreu81 is an extension for Stable Diffusion WebUI that enhances text-to-image generation with support for both English and Chinese languages, alongside img2img features. Although it has fewer stars (21) compared to others, its steady growth score of 6.25 suggests continuous improvement and user adoption.
video-diffusion-avatars by SimonGiebenhain provides code for turning video diffusion models into controllable 3D avatars. With a lower growth score but still significant interest (60 stars), this tool appeals to developers interested in advanced avatar creation techniques.
gemini-3-pro-image-preview-official-api-ja by apimarte80m2k offers an API for generating image previews using the Gemini 3 Pro model, with detailed documentation and examples. This project has garnered 80 stars and a steady growth score of 3.34, reflecting its utility in applications requiring fast image generation.
HarnessIR by PolyU-VCLab aims to restore real-world images using multimodal foundation models. With 34 stars and a moderate growth score of 2.96, it continues to attract interest from researchers and developers focused on image restoration techniques.
Lastly, harshpreet-hand by harshpreet931 showcases the process of converting handwritten notes into digital fonts through AI assistance. Despite limited recent activity (only one commit in a month), this project has attracted 56 stars due to its innovative approach to digitizing handwriting and transforming it into usable fonts.
These projects highlight the diversity and innovation within the Image & Video Generation space, catering to both broad and niche user bases with varying levels of engagement and technological advancement.
The tool dsh-video-storyboard by SAOS-3624 is designed for generating scene-aware storyboards in a single image format that includes timestamps, transcripts, and audio envelopes. This repository has seen significant growth with a score of 78.00, likely due to its comprehensive approach to visualizing video content in a concise manner.
adult-ai-video by wlnhnoqefoxt is an AI-driven generator for creating adult-themed videos from text descriptions and images. With a high growth score of 60.80 and over 200 stars, this project has attracted considerable attention, possibly due to its ability to generate NSFW content at a high resolution.
manga-generator by stockedge is a tool designed for manga creation, handling everything from scriptwriting to image generation and PDF output. It has gained moderate interest with 119 stars, indicating a strong community engagement among manga enthusiasts looking for AI-driven solutions to streamline the creative process.
pixel-loom by yangsir2077 is an open-source pixel art and animation editor that supports MCP-based AI creation of visual content. Despite having fewer stars at 89, its high commit activity (31 commits in the last month) suggests active development and user engagement within a dedicated community.
iris-3b by speridlabs is a diffusion transformer capable of direct pixel-space generation without VAE, along with fine-tuned capabilities for monocular depth estimation and image restoration. With 206 stars and a growth score of 7.35, this project stands out due to its versatile applications in computer vision tasks.
moss-image-forge-neo by eduardoabreu81 is an extension for Stable Diffusion WebUI that enhances text-to-image generation with support for both English and Chinese languages, alongside img2img features. Although it has fewer stars (21) compared to others, its steady growth score of 6.25 suggests continuous improvement and user adoption.
video-diffusion-avatars by SimonGiebenhain provides code for turning video diffusion models into controllable 3D avatars. With a lower growth score but still significant interest (60 stars), this tool appeals to developers interested in advanced avatar creation techniques.
gemini-3-pro-image-preview-official-api-ja by apimarte80m2k offers an API for generating image previews using the Gemini 3 Pro model, with detailed documentation and examples. This project has garnered 80 stars and a steady growth score of 3.34, reflecting its utility in applications requiring fast image generation.
HarnessIR by PolyU-VCLab aims to restore real-world images using multimodal foundation models. With 34 stars and a moderate growth score of 2.96, it continues to attract interest from researchers and developers focused on image restoration techniques.
Lastly, harshpreet-hand by harshpreet931 showcases the process of converting handwritten notes into digital fonts through AI assistance. Despite limited recent activity (only one commit in a month), this project has attracted 56 stars due to its innovative approach to digitizing handwriting and transforming it into usable fonts.
These projects highlight the diversity and innovation within the Image & Video Generation space, catering to both broad and niche user bases with varying levels of engagement and technological advancement.