Here we present a new method of estimating global variations in outdoor PM2.5 concentrations using satellite images combined with ground-level measurements and deep convolutional neural networks. Specifically, new deep learning models were trained over the global PM2.5 concentration range (<1-436 μg/m3) using a large database of satellite images paired with ground level PM2.5 measurements available from the World Health Organization. Final model selection was based on a systematic evaluation of well-known architectures for the convolutional base including InceptionV3, Xception, and VGG16. The Xception architecture performed best and the final global model had a root mean square error (RMSE) value of 13.01 μg/m3 (R2=0.75) in the disjoint test set. The predictive performance of our new global model (called IMAGE-PM2.5) is similar to the current state-of-the-art model used in the Global Burden of Disease study but relies only on satellite images as input. As a result, the IMAGE-PM2.5 model offers a fast, cost-effective means of estimating global variations in long-term average PM2.5 concentrations and may be particularly useful for regions without ground monitoring data or detailed emissions inventories. The IMAGE-PM2.5 model can be used as a stand-alone method of global exposure estimation or incorporated into more complex hierarchical model structures.