✳flâneur — a map of the web's best reading
One-Minute Video Generation with Test-Time Training
test-time-training.github.io · 19,783 words · saved by 1 readers
A new approach using Test-Time Training (TTT) layers to generate coherent, minute-long videos from text.
One-Minute Video Generation with Test-Time Training > One-Minute Demo Videos Click any video to see the text prompt used to generate it. The kitchen is tidy, featuring walls painted soft yellow and a checkered tile floor in white and pale blue. At the center stands a rectangular wooden kitchen table stained brown. Jerry, the brown mouse, sits upright on the kitchen table, holding a small block of yellow cheese in both paws. The camera remains steady, capturing Jerry's wide smile as he lifts the cheese toward his mouth and happily takes a large bite. The kitchen is tidy, featuring walls painted
Explore this link on the map →related reading
- Home | Genmogenmo.ai
- LLM Visualizationbbycroft.net
- The First Fully General Computer Action Model | blogsi.inc
- Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutationarxiv.org
- Swishy - AI Motion Designer & Animator | Create Stunning Animationsswishy.ai
- Arcads - Create winning ads with AIarcads.ai
- 2305.07759arxiv.org
- Replicate - Run AI with an APIreplicate.com
- GitHub - ml5js/training-styletransfer: Style Transfer training and using the model in ml5js · GitHubgithub.com
- Word to Worldxinyue.de
- GitHub - NVlabs/AutoGaze: AutoGaze automatically removes redundant patches in a video, reducing #tokens in ViT/MLLM by 4x-100x. · GitHubgithub.com
- Google Flow - AI Creative Studio for Video, Images & Custom Toolslabs.google