The goal of the Kinetics dataset is to help the computer vision and machine learning communities advance models for video understanding. Given this large human action classification dataset, it may be possible to learn powerful video representations that transfer to different video tasks.
The Kinetics-700-2020 dataset will be used for this challenge. Kinetics-700-2020 is a large-scale, high-quality dataset of YouTube video URLs which include a diverse range of human focused actions. The aim of the Kinetics dataset is to help the machine learning community create more advanced models for video understanding. It is an approximate super-set of both Kinetics-400, released in 2017, Kinetics-600, released in 2018 and Kinetics-700, released in 2019.
The dataset consists of approximately 650,000 video clips, and covers 700 human action classes with at least 700 video clips for each action class. Each clip lasts around 10 seconds and is labeled with a single class. All of the clips have been through multiple rounds of human annotation, and each is taken from a unique YouTube video. The actions cover a broad range of classes including human-object interactions such as playing instruments, as well as human-human interactions such as shaking hands and hugging.
More information about how to download the Kinetics dataset is available here.
Phoenix 1.5 RC2 represents a significant advancement in the realm of high-quality AI-driven content generation, specifically within the Leonardo.Ai Phoenix ecosystem. This release candidate (RC2) focuses on pushing the boundaries of photorealism, textural accuracy, and user control. Core Features of Phoenix 1.5 RC2
While earlier versions like Phoenix 1.4 or older RC builds focused on basic image generation, 1.5 RC2 emphasizes . This allows users to "iterate instantly" by refining a specific idea rather than starting from scratch, a feature that significantly improves the workflow for professional creators. Technical Integration Phoenix 1.5 Rc2 High Quality
: The update includes improved handling of reflective water pools and atmospheric lighting, contributing to the overall cinematic feel of the output. Comparison with Previous Iterations Phoenix 1
: The model excels at capturing fine details, such as the translucency of berries or the complex foliage of trees, creating a "dreamlike yet natural aesthetic". This allows users to "iterate instantly" by refining
The Phoenix 1.5 RC2 model is designed to produce high-fidelity imagery that rivals professional photography. Its primary improvements include:
: Users can describe changes in plain language, and Phoenix 1.5 RC2 updates the image in seconds, bypassing traditional technical steps.
For those coming from the world of RC flight simulation, "Phoenix" may also refer to the legacy Phoenix RC Simulator , which, although discontinued officially, remains active through community-driven updates. However, in the context of modern "High Quality" software keywords, the Leonardo.Ai Phoenix platform is currently the primary driver of this specific versioning. 5 RC2 generations? Phoenix - Leonardo.Ai
1. Possible to use ImageNet checkpoints?
We allow finetuning from public ImageNet checkpoints for the supervised track -- but a link to the specific checkpoint should be provided with each submission.
2. Possible to use optical flow?
Flow can be used as long as not trained on external datasets, except if they are synthetic.
3. Can we train on test data without labels (e.g. transductive)?
No.
4. Can we use semantic class label information?
Yes, for the supervised track.
5. Will there be special tracks for methods using fewer FLOPs / small models or just RGB vs RGB+Audio in the self-supervised track?
We will ask participants to provide the total number of model parameters and the modalities used and plan to create special mentions for those doing well in each setting, but not specific tracks.