Method for depth estimation for a variable focus camera

A computer-implemented method for extracting depth information from a plurality of images taken by a camera at different focus positions, the method comprising:

• capturing a sequence of images of a scene with the camera at different focus positions according to a predetermined focus schedule that specifies a chronological sequence of focus positions of the camera,
• extracting, by a machine learning algorithm comprising a convolutional neural network, image features of a predetermined number of captured images, wherein said extracted image features are stored individually for each of the predetermined number of captured images,
• after having extracted and stored image features from said predetermined number of captured images, processing, by the machine learning algorithm, a captured image whose image features have not yet been extracted, said captured image representing a currently processed image,
• said processing comprising extracting by the machine learning algorithm image features from the currently processed image and storing the extracted image features from the currently processed image, wherein, while performing the processing of the currently processed image, information of at least some or all of the other captured images is saved, re-used, and taken into account,
• said processing further comprising aligning image features stored from previously captured images with the image features of the currently processed image, and
• generating at least one multi-dimensional tensor representing the image features of at least some of the previously captured and processed images aligned to the image features of the currently processed image,
• generating a two-dimensional depth map using the focus positions specified in the predetermined focus schedule and the at least one generated multi-dimensional tensor.