Spatial transformer modules

1. A system comprising one or more computers and one or more storage devices storing instructions that when executed by the one or more computers cause the one or more computers to perform second operations comprising:

receiving one or more input images; and

processing the one or more input images using an image processing neural network system to generate a neural network output from the one or more input images, wherein the image processing neural network system comprises a spatial transformer module, and wherein during the processing of the one or more input images the spatial transformer module is configured to perform first operations comprising:

receiving an input feature map derived from the one or more input images, and

applying a spatial transformation to the input feature map to generate a transformed feature map, comprising:

processing the input feature map to generate, based on the input feature map, spatial transformation parameters that define the spatial transformation to be applied to the input feature map, and

sampling from the input feature map in accordance with the spatial transformation parameters generated based on the input feature map to generate the transformed feature map.

2. The system of claim 1 , wherein processing the one or more input images using the image processing neural network system comprises:

providing the transformed feature map as input to another component of the image processing neural network system.

3. The system of claim 1 , wherein the input feature map is an output generated by another component of the image processing neural network system.

4. The system of claim 3 , wherein the other component of the image processing neural network system is another spatial transformer module.

5. The system of claim 4 , wherein the other spatial transformer module performs a different type of spatial transformation than the spatial transformer module.

6. The system of claim 3 , wherein the other component of the image processing neural network system is a neural network layer.

7. The system of claim 1 , wherein the input feature map is one of the one or more input images.

8. The system of claim 1 ,

wherein the spatial transformer module comprises:

a localisation subnetwork comprising one or more neural network layers, wherein the localisation subnetwork is configured to process the input feature map to generate the spatial transformation parameters in accordance with current values of a set of parameters of the localisation subnetwork, and

wherein processing the input feature map to generate the spatial transformation parameters comprises processing the input feature map using the localisation subnetwork.

9. The system of claim 1 , wherein sampling from the input feature map in accordance with the spatial transformation parameters to generate the transformed feature map comprises:

generating, using the transformation parameters, a sampling grid that defines, for each of a plurality of locations in the transformed feature map, how a value of the location should be derived from values in the input feature map; and

sampling from the input feature map in accordance with the sampling grid to generate the transformed feature map.

10. The system of claim 1 , wherein the sampling mechanism is differentiable.

11. The system of claim 10 , wherein the spatial transformer module has been trained using backpropagation during training of the image processing neural network system.

12. The system of claim 1 , wherein the transformed feature map has the same dimensions as the input feature map.

13. The system of claim 1 , wherein the transformed feature map has different dimensions from the input feature map.

14. A method comprising:

receiving one or more input images; and

processing the one or more input images using an image processing neural network system to generate a neural network output from the one or more input images, wherein the image processing neural network system comprises a spatial transformer module, and wherein during the processing of the one or more input images the spatial transformer module is configured to perform operations comprising:

receiving an input feature map derived from the one or more input images, and

applying a spatial transformation to the input feature map to generate a transformed feature map, comprising:

processing the input feature map to generate, based on the input feature map, spatial transformation parameters that define the spatial transformation to be applied to the input feature map, and

sampling from the input feature map in accordance with the spatial transformation parameters generated based on the input feature map to generate the transformed feature map.

15. The method of claim 14 ,

wherein the spatial transformer module comprises:

a localisation subnetwork comprising one or more neural network layers, wherein the localisation subnetwork is configured to process the input feature map to generate the spatial transformation parameters in accordance with current values of a set of parameters of the localisation subnetwork, and

wherein processing the input feature map to generate the spatial transformation parameters comprises processing the input feature map using the localisation subnetwork.

16. The method of claim 14 , wherein sampling from the input feature map in accordance with the spatial transformation parameters to generate the transformed feature map comprises:

generating, using the transformation parameters, a sampling grid that defines, for each of a plurality of locations in the transformed feature map, how a value of the location should be derived from values in the input feature map; and

sampling from the input feature map in accordance with the sampling grid to generate the transformed feature map.

17. The method of claim 14 , wherein the sampling mechanism is differentiable.

18. One or more non-transitory computer readable storage media encoded with instructions that when executed by one or more computers cause the one or more computers to perform second operations comprising:

receiving one or more input images; and

processing the one or more input images using an image processing neural network system to generate a neural network output from the one or more input images, wherein the image processing neural network system comprises a spatial transformer module, and wherein during the processing of the one or more input images the spatial transformer module is configured to perform first operations comprising:

receiving an input feature map derived from the one or more input images, and

applying a spatial transformation to the input feature map to generate a transformed feature map, comprising:

processing the input feature map to generate, based on the input feature map, spatial transformation parameters that define the spatial transformation to be applied to the input feature map, and

sampling from the input feature map in accordance with the spatial transformation parameters generated based on the input feature map to generate the transformed feature map.

19. The computer readable storage media of claim 18 ,

wherein the spatial transformer module comprises:

a localisation subnetwork comprising one or more neural network layers, wherein the localisation subnetwork is configured to process the input feature map to generate the spatial transformation parameters in accordance with current values of a set of parameters of the localisation subnetwork, and

wherein processing the input feature map to generate the spatial transformation parameters comprises processing the input feature map using the localisation subnetwork.

20. The computer readable storage media of claim 18 , wherein sampling from the input feature map in accordance with the spatial transformation parameters to generate the transformed feature map comprises:

generating, using the transformation parameters, a sampling grid that defines, for each of a plurality of locations in the transformed feature map, how a value of the location should be derived from values in the input feature map; and

sampling from the input feature map in accordance with the sampling grid to generate the transformed feature map.