关键词 "still" 的搜索结果， 共 17 条, 只显示前 480 条
AI-powered platform for personalized learning
An open-source text-to-image generator using knowledge distillation.
Build your own AI assistant or chatbot with ChatNode's ChatGPT training platform.
Transform your photos into iconic movie stills and become a movie star with FilmFace AI App.
Transform still photos into lively expressive videos.
Analyze and distill ideas from launched products.
Generate high-quality images with SDXL Turbo.
Effortlessly extract still frames from videos for ChatGPT-4V.
AI-powered Uncoop platform reads and summarizes books, eliminating the need for reading.
Simplified, efficient content reading with AI.
Find perfect images for creative ideas.
Ultra-fast, high-quality AI image generation using ADD technology.
AI animation for still photos
The project is still in the incubation phase.
An ETH-obsessed MCP Server, where the creator is still pondering its ultimate usage. For now, it's just for fun!
ViLAMP（VIdeo-LAnguage Model with Mixed Precision）是蚂蚁集团和中国人民大学联合推出的视觉语言模型，专门用在高效处理长视频内容。基于混合精度策略，对视频中的关键帧保持高精度分析，显著降低计算成本提高处理效率。ViLAMP在多个视频理解基准测试中表现出色，在长视频理解任务中，展现出显著优势。ViLAMP能在单张A100 GPU上处理长达1万帧（约3小时）
MirageLSD 是 Decart AI 团队推出的全球首个 Live-Stream Diffusion（实时流扩散）AI 视频模型，能实现无限时长的实时视频生成，延迟低至 40 毫秒以内，支持 24 帧/秒的流畅输出。通过 Diffusion Forcing 技术和历史增强训练，解决了传统自回归模型在长时间生成中的误差累积问题，实现了视频的无限生成。基于Hopper 优化的 Mega Kern
只显示前20页数据，更多请搜索
