# Making CNNs for Video Parsing Accessible

## TL;DR

This paper presents methods to make CNN-based video parsing for e-sports more accessible by reducing computational requirements, enabling smaller organizations to extract game events from gameplay videos efficiently.

## Contribution

It introduces techniques to accelerate CNN training and inference, broadening access to automated game log extraction for non-expert users.

## Key findings

- Our approach outperforms standard backpropagation baselines.
- It enables training and prediction on less powerful hardware.
- The methods are validated on DOTA2 gameplay videos.

## Abstract

The ability to extract sequences of game events for high-resolution e-sport games has traditionally required access to the game's engine. This serves as a barrier to groups who don't possess this access. It is possible to apply deep learning to derive these logs from gameplay video, but it requires computational power that serves as an additional barrier. These groups would benefit from access to these logs, such as small e-sport tournament organizers who could better visualize gameplay to inform both audience and commentators. In this paper we present a combined solution to reduce the required computational resources and time to apply a convolutional neural network (CNN) to extract events from e-sport gameplay videos. This solution consists of techniques to train a CNN faster and methods to execute predictions more quickly. This expands the types of machines capable of training and running these models, which in turn extends access to extracting game logs with this approach. We evaluate the approaches in the domain of DOTA2, one of the most popular e-sports. Our results demonstrate our approach outperforms standard backpropagation baselines.