date
type
status
slug
summary
tags
category
icon
password
Beyond Appearances - CVPR AI Art Gallery 2024 artwork (with Tutorial)
— HALF FOR THE AI, HALF RESERVED FOR MYSELF
Using a smartphone (Xiaomi 12T Pro) attached to an external microscope, I captured microcosmic plant images that reveal intricate details invisible to the naked eye. These images showcase the complexity and beauty of nature at a microscopic level.
I shared half of these glimpses with AI, allowing it to process and interpret them, uncovering patterns and aspects that might be overlooked by human observation. The other half of the images were kept for my personal reflection, maintaining a connection with the raw, unaltered essence of the captured moments. These serve as a canvas for my own interpretations, grounded in human experience and emotion.
The unseen remnants are left to the flowing sands of time to unveil, suggesting that some aspects of reality will always remain elusive and mysterious. This project, “Beyond Appearances,” blends AI interpretation with personal insight, challenging conventional views of reality and encouraging exploration of deeper layers of meaning within the natural world.
➡️Link to CVPR AI Art Gallery :https://thecvf-art.com/project/shengyu-meng/
AI Creation Workflow
- Capture Original Photos • Take microcosmic original photos as the base material.
2. Recreated the Photos Using Stable Diffusion
• Repaint the original photos using Stable Diffusion to reveal intricate hidden details.
3. Combine Photos
• Manually blend the original photos with the Stable Diffusion repainted output.
4. Create Video Using Stable Video Diffusion
• Apply Stable Video Diffusion to convert the still image to video.
5. Frame Interpolation for Smooth Transitions
• Use frame interpolation to fuse different keyframes, ensuring smooth transitions between scenes.
6. Edit and Compile in Capcut
• Combine all AI-generated footages and edit the video using Capcut .
7. Post-Process Video
• Enhance the video further with AI super resolution and additional frame interpolation to achieve the best possible quality.
- 作者:Simon Shengyu Meng
- 链接:https://simonsy.net/article/CVPR-2024
- 声明:本文采用 CC BY-NC-SA 4.0 许可协议,转载请注明出处。
相关文章
DreamGaussian: The Stable Diffusion Moment of AIGC 3D Generation
How I Used AI to Create a Promotional Video for Xiaomi's Daniel Arsham Limited Edition Smartphone
Works Series - Dimensional Recasting
The 2022 Venice - Metaverse Art Annual Exhibition: How Nature Inspires Design
Hidden Time Space 5: Cloudscape Artistry — "Vintage" AI Model Generates Retro Chinese-Style Landscape Animation
The Basic Principles of ChatGPT