To attain speed estimation, the OHL pixels are engineered in such a way that their VOC output is dependent on illumination time. This approach is based on the concept that light from fast objects dwell on the corresponding pixels for a shorter time than slower ones. As Figure 2a illustrates, the pixel’s VOC slowly rises with illumination time. This analog VOC response provides temporal information about the detected motion, i.e. at a given light intensity, the VOC pixel values reveal the speed of the moving object. To demonstrate this type of object tracking, for simplicity, we have limited ourselves to one dimensional motion first. Expansion to two-dimensional motion can be achieved by integrating more pixels. Figure 5f depicts the VOC values of each pixel registering a 2 x 2 pixel white square moving from left to right at a uniform frame rate of 50 ms (movie 4). Since the motion is uniform in our experiment, the VOC responses of each pixel should be similar at a given speed. Furthermore, various constant speeds should result in different VOC corresponding to the analog VOC dependence on light exposure time. This expected behavior is in alignment with our results, as illustrated in Figure 5f, with faster motion yielding lower VOC values (statistical interpretation can be seen on Figure 5i). However, it is to be noted that the OHL pixels themselves cannot correctly perform speed estimation since one VOC value may represent different speeds (illumination times) depending on the registered light intensity (Figure 2a). Hence, to properly assess the speed, both VOC (from OHL) and light intensity values (from ODL) have to be known. This requirement is satisfied by the double layered sensor architecture (Figure 1a). Therefore, our sensor, in addition to its self-learning capabilities, also allows very straightforward speed estimation by simply probing both the ODL and OHL pixels without having to measure time between successive image frames representing the object’s motion, leading to significantly reduced computational costs.