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.
The HD version of On the Basis of Sex captures the tension of this argument brilliantly. In close-up, you can see Felicity Jones’s Ginsburg trembling slightly as she approaches the bench—not from fear, but from the weight of knowing that she is arguing for a principle that would ripple across generations.
Elena Vasquez was the best junior litigator at Hale, Crenshaw & Finch. She had argued three motions, won two dismissals, and drafted the brief that saved their largest corporate client. But when the partnership track was announced, her name wasn't on it. on the basis of sexhd
The film focuses on RBG’s early legal battles, specifically (1972). She argues that a section of the tax code discriminates based on sex — a man was denied a caregiver tax deduction simply because he was male. This case was her strategic first step to prove that gender-based laws harm everyone , not just women. Beyond the trope: Why “Based on a True
Buffering is the enemy of HD streaming. To watch without interruption, you need sufficient bandwidth. She had argued three motions, won two dismissals,
Ruth Bader Ginsburg’s early lawsuits targeted laws that treated men and women differently—showing how discriminatory statutes harmed both sexes. The film condenses and dramatizes events, but its core—challenging legal norms to expand constitutional protections—reflects real legal strategies used by Ginsburg and the ACLU Women’s Rights Project.
: Critics from platforms like Variety and The Crimson describe it as an "inspiring origin story" that humanizes a civil rights icon while occasionally leaning into "feminist iconography". Where to Watch
On the Basis of Sex reminds audiences that constitutional law evolves through determined advocacy, careful argument, and sometimes unlikely alliances. It offers an accessible entry point to discussions about sex discrimination, equal protection, and how cultural change and legal reform interact.
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.