OctWave 3.0 – Kaggle Challenge - Round 2
August 16, 2026 at 6:29 PM - August 17, 2026 at 6:29 PM @ Online event
Description
The second round of the OctWave 3.0 Kaggle Challenge was conducted virtually from 16th August 2026 at 11:59 PM to 17th August 2026 at 11:59 PM, featuring the top 70 teams qualified from Round 1. This stage increased the technical difficulty of the competition by presenting participants with a challenging computer vision problem focused on image classification. At the commencement of the 24-hour challenge, the competition dataset and participation guidelines were released. Participants competed either individually or in teams of up to four members and were tasked with developing deep learning models to accurately detect and classify character appearances from image frames. To maintain fairness and integrity, participants were required to follow the official team-naming conventions, restrictions on external private datasets, and code-sharing regulations. During the challenge, the 70 qualified teams developed computer vision pipelines using techniques such as image preprocessing, data augmentation, and advanced deep learning architectures. Models were evaluated using the Macro F1-score, allowing performance to be assessed appropriately across imbalanced classes. Following the conclusion of the Kaggle competition, a verification phase was conducted on 18th August 2026. Participants were required to submit their complete Jupyter Notebook and technical documentation through the designated submission portal within an approximately six-hour submission window. The organizing committee subsequently evaluated the leaderboard rankings, reproducibility of the submitted code, and technical methodology. Following this evaluation, the top 10 teams were selected to advance to the Grand Finale of OctWave 3.0. Round 2 provided participants with an opportunity to apply advanced computer vision and deep learning techniques to a challenging practical problem while demonstrating their technical knowledge, problem-solving ability, and model-development skills.
Agenda
16 August 2026 – 11:59 PM Round 2 commenced with the release of the Kaggle competition dataset, participation guidelines, and competition rules. 16–17 August 2026 Qualified teams developed and evaluated deep learning models for the computer vision image-classification task. Participants performed preprocessing, image augmentation, model training, and optimization, with submissions evaluated using the Macro F1-score. 17 August 2026 – 11:59 PM Kaggle Challenge Round 2 concluded. 18 August 2026 Verification and documentation submission phase was conducted. Teams submitted their Jupyter Notebooks and technical documentation through the designated submission portal. Following evaluation Leaderboard performance, reproducible code, and methodology reports were reviewed by the organizing committee, and the top 10 teams were selected to advance to the Grand Finale.