TWI716926B

An object posture recognition method suitable for implementing on a photographed image is provided. The photographed image shows a first object and a second object. The object posture recognition method comprises the steps of: (A) recognizing a plurality of critical areas corresponding to the objects from the photographed image by an object posture recognition system, and performing the classification process of the critical areas, the object posture recognition system also performs feature extraction process of the photographed image to generate a plurality of feature data respectively corresponding to the critical areas; (B) generating a first posture estimation result and a second posture estimated result by the object posture recognition system based upon a first three-dimensional point cloud model, a second three-dimensional point cloud model, and the feature data, the first and second posture estimation results respectively indicating the postures of the first and second objects in the three-dimensional