A computer vision method is presented for recognizing the non-rigid motion observed in objects moving in a 3D environment. This method is embedded in a more complete mechanism that integrates low-level (image processing), mid- level (recursive 3D trajectory estimation), and high-level (action recognition) processes. Multiple moving objects are observed via a single, uncalibrated video camera. A Kalman filter formulation is used in estimating the relative 3D motion trajectories. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages. In this paper we concentrate in the action recognition stage. The 3D trajectory, occlusion, and segmentation information are utilized in extracting stabilized views of the moving object. Trajectory-guided recognition (TGR) is then proposed as an efficient method for adaptive classification of action. The TGR approach is demonstrated using 'motion history images' that are then recognized via a mixture of Gaussian classifier. The system was tested in recognizing various dynamic human outdoor activities; e.g., running, walking, roller blading, and cycling.