The Spitzer Survey of Stellar Structure in Galaxies (S$^4$G, Sheth et. al. 2010) is a deep 3.6 and 4.5 $mu$m imaging survey of 2352 nearby ($< 40$ Mpc) galaxies. We describe the S$^4$G data analysis pipeline 4, which is dedicated to 2-dimensional str
uctural surface brightness decompositions of 3.6 $mu$m images, using GALFIT3.0 citep{peng2010}. Besides automatic 1-component Sersic fits, and 2-component Sersic bulge + exponential disk fits, we present human supervised multi-component decompositions, which include, when judged appropriate, a central point source, bulge, disk, and bar components. Comparison of the fitted parameters indicates that multi-component models are needed to obtain reliable estimates for the bulge Sersic index and bulge-to-total light ratio ($B/T$), confirming earlier results citep{laurikainen2007, gadotti2008, weinzirl2009}. In this first paper, we describe the preparations of input data done for decompositions, give examples of our decomposition strategy, and describe the data products released via IRSA and via our web page ({bf tt www.oulu.fi/astronomy/S4G_PIPELINE4/MAIN}). These products include all the input data and decomposition files in electronic form, making it easy to extend the decompositions to suit specific science purposes. We also provide our IDL-based visualization tools (GALFIDL) developed for displaying/running GALFIT-decompositions, as well as our mask editing procedure (MASK_EDIT) used in data preparation. In the second paper we will present a detailed analysis of the bulge, disk, and bar parameter derived from multi-component decompositions.
