Skip to content
Hot storyWatching

Multi-label aware game bug detection model and dataset released

1 report1 sourceUpdated 3 d ago

What happened

AI overview

Nahian Rifaat and colleagues propose a ResNet-BiLSTM model that performs multi-label perceptual bug detection from gameplay footage, reaching an F1 score of 85.78% on a benchmark dataset and comparing against video classification models such as Inflated 3D ConvNet and 3D ResNet. The multi-label perceptual bug dataset released alongside it contains 77,969 video clips and roughly 1.2 million frames, covers different game genres, and allows combinations of 5 bug categories to appear within the same frame.

AI-written from the coverage · updated 2 h ago

LatestOct 7 00:01
Study proposes ResNet-BiLSTM for multi-label perceptual bug detection from gameplay footage

Coverage timeline

Follow the coverage to see the story from different sides.

Oct 7
  1. arXiv Game AIPick
    Study proposes ResNet-BiLSTM for multi-label perceptual bug detection from gameplay footage

    The study proposes a ResNet-BiLSTM model that performs multi-label perceptual bug detection on gameplay footage, achieving an F1 score of 85.78% on a benchmark dataset, and compares it with video classification models including Inflated 3D ConvNet and 3D ResNet.

Heat trend for this story

Not enough continuous observations to chart a trend yet.