This paper presents simple and effective baseline methods for human pose estimation and tracking, aiming to inspire new ideas and simplify evaluation in the field. The authors achieve state-of-the-art results on challenging benchmarks, including the COCO Keypoint Challenge and the PoseTrack Challenge. For pose estimation, a deconvolutional head network is used, which adds a few deconvolutional layers to a ResNet backbone. For pose tracking, a flow-based approach is proposed, which uses optical flow to propagate joints across frames and a flow-based pose similarity metric for tracking. The code and pre-trained models are available online. The paper includes comprehensive ablation studies and comparisons with state-of-the-art methods, demonstrating the effectiveness and simplicity of the proposed baselines.This paper presents simple and effective baseline methods for human pose estimation and tracking, aiming to inspire new ideas and simplify evaluation in the field. The authors achieve state-of-the-art results on challenging benchmarks, including the COCO Keypoint Challenge and the PoseTrack Challenge. For pose estimation, a deconvolutional head network is used, which adds a few deconvolutional layers to a ResNet backbone. For pose tracking, a flow-based approach is proposed, which uses optical flow to propagate joints across frames and a flow-based pose similarity metric for tracking. The code and pre-trained models are available online. The paper includes comprehensive ablation studies and comparisons with state-of-the-art methods, demonstrating the effectiveness and simplicity of the proposed baselines.