FiLM: Visual Reasoning with a General Conditioning Layer

TL;DR

FiLM introduces a versatile conditioning layer that significantly improves neural network performance in visual reasoning tasks by enabling effective feature modulation, leading to state-of-the-art results and strong generalization.

Contribution

The paper presents FiLM, a novel feature-wise linear modulation technique that enhances neural networks' ability to perform complex visual reasoning tasks.

Findings

Halves state-of-the-art error on CLEVR benchmark

Effectively modulates features for reasoning tasks

Generalizes well to new data with few or zero examples

Abstract

We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. We show that FiLM layers are highly effective for visual reasoning - answering image-related questions which require a multi-step, high-level process - a task which has proven difficult for standard deep learning methods that do not explicitly model reasoning. Specifically, we show on visual reasoning tasks that FiLM layers 1) halve state-of-the-art error for the CLEVR benchmark, 2) modulate features in a coherent manner, 3) are robust to ablations and architectural modifications, and 4) generalize well to challenging, new data from few examples or even zero-shot.