3D vision processing using an IP block

1. A system for 3D vision processing, the system comprising:

a sensor device;

a processor;

a memory module;

a network interface; and

a vision processing module comprising an integrated circuit that performs one or more 3D vision processing algorithms, including simultaneous localization and mapping (SLAM) algorithms comprised of a Build Pyramid function, a Key Point Detection function, and a Feature Calculation function, and a plurality of controllers that couple the integrated circuit to each of: the sensor device, the processor, the memory module, and the network interface;

wherein:

the vision processing module receives image data from the sensor device, the image data corresponding to one or more images captured by the sensor device;

the vision processing module executes one or more of the 3D vision processing algorithms using at least a portion of the image data as input, including executing the Build Pyramid function to:

receive color and depth data and a timestamp associated with the color and depth data from the sensor device;

generate a gray map comprising a plurality of levels, using the color and depth data and the timestamp;

resize the gray map;

smooth a first level of the plurality of levels of the resized gray map; and

determine whether to smooth all of the plurality of resized levels or to resize the smoothed first level; and

the vision processing module transmits an output from execution of one or more of the 3D vision processing algorithms to at least one of: the processor, the memory module, or the network interface.

2. The system of claim 1 , wherein the vision processing module comprises a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC).

3. The system of claim 1 , wherein when executing the Key Point Detection function, the vision processing module:

executes a features from accelerated segment test (FAST) corner detection algorithm with non-maxima suppression on the resized gray map to detect one or more key points in the resized gray map; and

determines whether one or more key points have been detected in each of the plurality of levels in the resized gray map.

4. The system of claim 3 , wherein when executing the Feature Calculation function, the vision processing module:

performs moment calculation on each detected key point to generate M01 and M10 values for the key point;

calculate a magnitude value for each key point using the corresponding M01 and M10 values;

rotate trained pattern coordinates for each key point based upon the M01 and M10 values and the magnitude for the corresponding key point; and

calculates final ORB values using the rotated trained pattern coordinates.

5. The system of claim 4 , wherein the vision processing module transmits the final ORB value, the M01 value, the M10 value, and the magnitude value of each key point to the processor, and the processor generates an intensity angle for the corresponding key point based upon the magnitude value, the M01 value and the M10 value of the key point.

6. The system of claim 5 , wherein the vision processing module comprises one instance of each of the Build Pyramid function, the Key Point Detection function, and the Feature Calculation function.

7. The system of claim 5 , wherein the vision processing module comprises a plurality of instances of each of the Build Pyramid function, the Key Point Detection function, and the Feature Calculation function.

8. The system of claim 1 , wherein the processor, the memory module, the sensor device and the vision processing module are embedded in a System-on-Chip (SoC).

9. The system of claim 8 , wherein the SoC is embedded in a mobile computing device.

10. The system of claim 8 , wherein the processor, the memory module, the sensor device and the vision processing module are coupled to each other via a system bus.

11. The system of claim 1 , wherein the network interface couples the vision processing module to a remote computing device.

12. The system of claim 1 , wherein at least a portion of one or more of the 3D vision processing algorithms is executed by the processor in coordination with the vision processing module.

13. The system of claim 1 , wherein the sensor device is a camera.