Camera system and method of controlling a computing hardware accelerator

1 . A computer-implemented method of controlling a computing hardware accelerator in a movable camera, the method comprising:

acquiring images from the movable camera;

obtaining an indication of a physical state of the movable camera, indicating whether the movable camera is in a moving state or in a still state;

allocating exclusive access to the computing hardware accelerator based on the indication of the physical state of the movable camera, wherein:

if the movable camera is in the moving state, allocating the computing hardware accelerator to a first processing workload related to the acquired images and associated with the moving state; and

if the movable camera is in the still state, allocating the computing hardware accelerator to a second processing workload related to the acquired images and associated with the still state, wherein the first processing workload and the second processing workload are mutually exclusive with respect to the physical state of the movable camera and are each accelerated by the same computing hardware accelerator.

2 . The method of claim 1 , wherein the first processing workload comprises computation of graphics in the images acquired by the movable camera.

3 . The method of claim 1 , wherein the first processing workload comprises computation of one or more privacy masks in the images acquired by the movable camera.

4 . The method of claim 3 , wherein the computation of one or more privacy masks comprises dynamically computing an area and/or a selection of pixels and/or polygons to be blurred and/or covered and/or anonymized in the images acquired by the movable camera.

5 . The method of claim 1 , wherein the second processing workload comprises analytics computations related to the images acquired by the movable camera.

6 . The method of claim 5 , wherein the analytics computations comprise computations related to object detection in the images acquired by the movable camera.

7 . The method of claim 1 , wherein the second processing workload comprises computations related to augmented reality content to be added to the images acquired by the movable camera.

8 . The method of claim 1 , wherein the second processing workload comprises applying a machine learning model trained to perform a specific image and/or video analysis workload on the images acquired by the movable camera.

9 . The method of claim 1 , wherein the indication of the physical state of the movable camera is obtained continuously or at intervals.

10 . The method of claim 1 , further comprising the step of automatically switching between allocating the computing hardware accelerator to the first processing workload and allocating the computing hardware accelerator to the second processing workload based on the indication of the physical state.

11 . The method of claim 1 , wherein the movable camera is a pan-tilt or a pan-tilt-zoom camera.

12 . The method of claim 1 , wherein the moving state of the movable camera indicates mechanical movement of the movable camera.

13 . The method of claim 1 , wherein the indication of the physical state further comprises information about a movement direction and/or movement speed of the movable camera, and wherein the first processing workload or the second processing workload is adapted based on the movement direction and/or movement speed.

14 . A non-transitory computer readable recording medium comprising a computer program having instructions which, when executed by a computing device or computing system, cause the computing device or computing system to carry out a method of controlling a computing hardware accelerator in a movable camera, the method comprising:

acquiring images from the movable camera;

obtaining an indication of a physical state of the movable camera, indicating whether the movable camera is in a moving state or in a still state;

allocating exclusive access to the computing hardware accelerator based on the indication of the physical state of the movable camera, wherein:

if the movable camera is in the moving state, allocating the computing hardware accelerator to a first processing workload related to the acquired images and associated with the moving state; and

if the movable camera is in the still state, allocating the computing hardware accelerator to a second processing workload related to the acquired images and associated with the still state, wherein the first processing workload and the second processing workload are mutually exclusive with respect to the physical state of the movable camera and are each accelerated by the same computing hardware accelerator.

15 . A camera system, comprising:

a movable camera;

a computing hardware accelerator; and

a central processing unit, wherein the central processing unit is configured to:

acquire images from the movable camera;

obtain an indication of a physical state of the movable camera, indicating whether the movable camera is in a moving state or in a still state;

allocate exclusive access to the computing hardware accelerator based on the indication of the physical state of the movable camera, wherein:

if the movable camera is in the moving state, allocate the computing hardware accelerator to a first processing workload related to the acquired images and associated with the moving state; and

if the movable camera is in the still state, allocate the computing hardware accelerator to a second processing workload related to the acquired images and associated with the still state, wherein the first processing workload and the second processing workload are mutually exclusive with respect to the physical state of the movable camera and are each accelerated by the same computing hardware accelerator.