F-CAM: Full Resolution Class Activation Maps via Guided Parametric Upscaling

TL;DR

This paper introduces F-CAM, a trainable parametric upscaling method for CAMs that significantly improves localization accuracy in weakly-supervised object localization tasks while reducing computational costs.

Contribution

The authors propose a novel trainable decoder architecture for full resolution CAMs that leverages image priors and fine-tuning, outperforming existing methods in accuracy and efficiency.

Findings

F-CAM achieves higher localization accuracy across multiple CNN backbones.

The method is computationally more efficient during inference.

F-CAM outperforms several state-of-the-art WSOL techniques on benchmark datasets.

Abstract

Class Activation Mapping (CAM) methods have recently gained much attention for weakly-supervised object localization (WSOL) tasks. They allow for CNN visualization and interpretation without training on fully annotated image datasets. CAM methods are typically integrated within off-the-shelf CNN backbones, such as ResNet50. Due to convolution and pooling operations, these backbones yield low resolution CAMs with a down-scaling factor of up to 32, contributing to inaccurate localizations. Interpolation is required to restore full size CAMs, yet it does not consider the statistical properties of objects, such as color and texture, leading to activations with inconsistent boundaries, and inaccurate localizations. As an alternative, we introduce a generic method for parametric upscaling of CAMs that allows constructing accurate full resolution CAMs (F-CAMs). In particular, we propose a trainable decoding architecture that can be connected to any CNN classifier to produce highly accurate CAM localizations. Given an original low resolution CAM, foreground and background pixels are randomly sampled to fine-tune the decoder. Additional priors such as image statistics and size constraints are also considered to expand and refine object boundaries. Extensive experiments, over three CNN backbones and six WSOL baselines on the CUB-200-2011 and OpenImages datasets, indicate that our F-CAM method yields a significant improvement in CAM localization accuracy. F-CAM performance is competitive with state-of-art WSOL methods, yet it requires fewer computations during inference.