Abstract: We develop a technique for generating smooth and accurate 3D human pose and motion estimates from RGB video sequences. Our technique, which we call Motion Estimation via Variational Autoencoder (MEVA), decomposes a temporal sequence of human motion into a smooth motion representation using auto-encoder-based motion compression and a residual representation learned through motion refinement. This two-step encoding process of human motion can represent a wide variety of general human motions while also retaining person-specific motion details. Experiments show that our method produces both smooth and accurate 3D human pose and motion estimates.