Seamounts, generally formed by volcanic activity, play a crucial role in marine ecosystems, ocean circulation, and plate tectonics. Cataloguing the world´s seamounts has usually relied on analysing vertical gravity gradient (VGG) data, in which edifices smaller than 2 km generally produce no signal, leaving a significant gap in our understanding of ocean floor topography. Most of the seamounts identified in VGG data have been picked by hand, in view of the >40.000 features presently identified, a very time-consuming process. With the Seabed 2030 initiative making ever-increasing amounts of multibeam data publicly accessible, it has become extremely important to leverage this detailed bathymetry data for seamount detection and combine it with computer-aided detection methods. We present a novel framework that automatically detects and catalogs seamounts in multibeam data, enhancing comprehensive mapping of seamount distribution across a range of edifice sizes.Our framework initially segments the multibeam data into 128x128 sized images, allowing it to manage large datasets. It then uses Convolutional Neural Networks (CNNs) to calculate feature vectors. These vectors are used to cluster segmented images, distinguishing flat seafloor areas from those with potential seamounts. Potential summits are flagged and then evaluated by calculating features such as slope, shape, and summit area, resulting in a detailed catalog of seamounts with information on their characteristics and sizes.The framework´s capability to automatically scan large datasets in a short time enhances efficiency of seamount detection. The framework is also extendable, allowing additional seafloor features, such as ridges or trenches, to be included. It holds significant potential for improving our understanding of ocean floor topography and seamount distribution and for analysing the terabytes of bathymetric data held in international repositories.The resulting catalogues allow us, among others, to address questions of the diverse causes of magmatism in the ocean basins, potentially quantify the influence of seamounts on deep ocean mixing and provide important fundamental data on the distribution of habitat types on the deep sea.