Overview
Who this competition fits
Advanced practitioners with experience in 3D image classification and security-domain applications. Requires comfort with large-scale image data and deep learning pipelines.
Read the original official blurb
Improve the accuracy of the Department of Homeland Security's threat recognition algorithms
Preparation
From registration to a first submission
- 01
Python
- 02
deep learning (CNNs)
- 03
3D image processing
- 04
large dataset handling
- 05
GPU training infrastructure
Before you commit: Uses 3D body scanner images which require specialized preprocessing. The dataset is very large, the problem involves multi-zone threat detection, and the security domain adds complexity to validation.
Source
How this page was assembled
Competition information is structured from the official page. Scores are platform estimates for decision support; official rules take precedence.
- Official competition page
- Kaggle
- Last checked
- Jun 22, 2017