Our group has a long expertise in Image processing in the framework of diverse applications (health, surveillance, Assistance Driving, etc). It has participated in more than 10 EU projects in the last 10 years and has a solid track record in scientific production in this field.
In our group we have long (very long and extensive expertise on low image processing, we have done optical flow algorithms (energy based, gradient based, block matching, etc) and similar approaches for stereo models, we also have models for background substraction (in surveillance applications) and local contrast descriptors (pixel wise orientation, magnitud, intrinsic dimensionality, etc). We have applied all this to ADAS (in vehicles), to surveillance applications, to robotics and to UAVs.
Some of our aproaches, we adapted the vision models for focal plane processors (or hibrid processing because they also used an embedded RISC) for instance with the system of Angel Rodriguez (in ANAFOCUS), among others, also with expertise in LVDS sensors (event driven retina like sensors). We also implemented the models in GPU, CPU, and FPGA.
Well, see our expertise in the list of publications bellow. Thus we can be involved in developing the models for different architecures (your target platform and also FPGA, CPU, GPU, etc). And perform performance comparison (benchmarking, accuracy vs efficiency trade offs).
In our group we have long (very long and extensive expertise on low image processing, we have done optical flow algorithms (energy based, gradient based, block matching, etc) and similar approaches for stereo models, we also have models for background substraction (in surveillance applications) and local contrast descriptors (pixel wise orientation, magnitud, intrinsic dimensionality, etc). We have applied all this to ADAS (in vehicles), to surveillance applications, to robotics and to UAVs.
Some of our aproaches, we adapted the vision models for focal plane processors (or hibrid processing because they also used an embedded RISC) for instance with the system of Angel Rodriguez (in ANAFOCUS), among others, also with expertise in LVDS sensors (event driven retina like sensors). We also implemented the models in GPU, CPU, and FPGA.
Well, see our expertise in the list of publications bellow. Thus we can be involved in developing the models for different architecures (your target platform and also FPGA, CPU, GPU, etc). And perform performance comparison (benchmarking, accuracy vs efficiency trade offs).
- Please, check here, a few slides for a summary of our expertise.