High speed imaging is a booming activity with the ideal application of CMOS technology imagers. It makes it possible to acquire a fast single event at a fast sampling and frame rate and to observe it at a reduced frame rate. It finds many applications in motion analysis, explosives, ballistic, biomechanics research, crash test, airbag deployment, manufacturing, production line monitoring, deformation, droplet formation, fluid dynamics, particle, spray, shock & vibration, etc. High speed video imaging is currently driven by some industrial manufacturers such as Photron, Redlake, Drs Hadland, which design their own sensors. The current industrial most efficient imagers offer a speed of 22,000 frames per second (fps) for a spatial resolution of 1280x800 pixels, i.e. 22 Gpixel/s. This speed is not restricted by the electronics of the pixel but by the sensors chip inputs/outputs interconnections. The conventional operation mode based on extracting the sensor data at each acquisition of a new image is a real technological barrier that limits the scope of high speed cameras to the study of transient phenomena that last for a few hundred microseconds. The FALCON project's main goal is to overcome this technological barrier, increasing the acquisition speed by three orders of magnitude by proposing a sensor capable of taking up to 100 million fps while increasing the sampling rate up to 10 TeraPixel/s. To accomplish this, the classical approach of extracting image sensor should be abandoned in favor of a new one which makes it possible to eradicate the inputs/outputs bottleneck. Several studies mention the realization of high-speed image sensors based on the principle of "burst" imagers (BIS Burst Image Sensor). Since it is impossible to get the frames out of the sensor as they are acquired, the idea is to store all the images in the sensor and execute the readout afterward, after the end of the event to be recorded. So far, all the developed BIS based on this principle use a totally analog approach in the form of a monolithic sensor. The size of the embedded memory is generally limited to a hundred frames, the pixel pitch is around 50 µm and the acquisition rate is in the order of 10 Mfps for large 2D arrays. Furthermore, research works mention little data about the signal to noise ratio (SNR), but the leakage current of the storage capacities degrades the signal quality and the effect is more noticeable when the readout duration is high, i.e. when the number of stored images is large. This phenomenon limits, once again, the number of storable images in analog BIS forms. In general, a maximal SNR of 45 dB is obtained. The FALCON project is based on a device concept in total disruption with previous works, by implementing the possibilities offered by the emergent microelectronics 3D technologies in order to increase the performance of this type of sensor while also adding more features to it. A PhD work started in 2012 in collaboration between the CEA Leti and the ICube laboratory helped to determine an optimal sensor architecture that takes advantage of the 3D technology. A particularity of the proposed architecture is the in-line analog to digital conversion at full speed. This study shows that the proposed new approach increases the number of stored images, while increasing the signal to noise ratio. It has also brought light to the potential problems of heat dissipation inherent to both fast circuits and 3D technologies. The methodological aspects of the design are also at the center of the project seeing that architecture/partitioning and electronic/thermal co-designs are necessary to carry out this type of conception. New tools and methods for the design of integrated heterogeneous systems are needed. The ultimate objective of the project is a high definition 1200x1200 pixels, 10 Mega fps with more than 1000 frames embedded digital memory. The project is pushing the performances of all the system bricks to the state of the art.