Starstruk Logo MarkStarstruk

v1.1.0

joshiayush/inbv1.1.0May 7, 2024by FrankLeeeee

AI Summary

This release expands Open-Sora's capabilities to handle longer videos and diverse resolutions while introducing an automated data processing pipeline and an improved ST-DiT architecture.

Key Highlights

  • Improved ST-DiT architecture with RoPE and QK norm
  • Automated data processing pipeline
  • Support for any resolution, aspect ratio, and duration
  • Image and video conditioning support
  • Video editing capabilities (animating images, connecting videos)

New Features

  • ST-DiT architecture improvements (RoPE, QK norm, longer text)
  • Automated data processing pipeline (scene cutting, filtering, captioning)
  • Variable resolution and duration support
  • Image-to-video and video-to-video editing

Full Release Notes

📍 Open-Sora 1.1 released
- 🌠 Model weights are available [here](https://github.com/hpcaitech/Open-Sora). It is trained on 0s~15s, 144p to 720p, various aspect ratios videos. See our [report 1.1](https://github.com/hpcaitech/docs/report_02.md) for more discussions.
- 🔧 Data processing pipeline v1.1 is released. An automatic [processing pipeline](https://github.com/hpcaitech/Open-Sora#data-processing) from raw videos to (text, video clip) pairs is provided, including scene cutting, filtering(aesthetic, optical flow, OCR, etc.), captioning managing. With this tool, you can easily build your video dataset.
- ✅ Improved ST-DiT architecture includes rope positional encoding, qk norm, longer text length, etc.
- ✅ Support training with any resolution, aspect ratio, and duration (including images).
- ✅ Support image and video conditioning and video editing, and thus support animating images, connecting videos, etc.

Visit the [Open-Sora Gallery](https://hpcaitech.github.io/Open-Sora/) to view more samples.