If we wanted to pass the full results data from a FOMO or any other classification would we just pass these structs?
typedef struct {
const char *label;
uint32_t x;
uint32_t y;
uint32_t width;
uint32_t height;
float value;
} ei_impulse_result_bounding_box_t;
typedef struct {
int sampling;
int dsp;
int classification;
int anomaly;
int64_t dsp_us;
int64_t classification_us;
int64_t anomaly_us;
} ei_impulse_result_timing_t;
typedef struct {
//#if EI_CLASSIFIER_OBJECT_DETECTION == 1
// ei_impulse_result_bounding_box_t bounding_boxes[EI_CLASSIFIER_OBJECT_DETECTION_COUNT];
ei_impulse_result_bounding_box_t bounding_boxes[10];
//#else
// ei_impulse_result_classification_t classification[EI_CLASSIFIER_LABEL_COUNT];
//#endif
float anomaly;
ei_impulse_result_timing_t timing;
int32_t label_detected;
} ei_impulse_result_t;
// The big result coming from the M7 core of the Portenta to the M4 core
ei_impulse_result_t resultM4;
I can pass serial data between cores using one of my examples here
Presently getting code to compile just no data is flowing. I will be doing some more testing.
Kind of related to this post Advanced Vision FOMO adding keras layers - #4 by Rocksetta
My github for this will be at here but the code is very messy as it has OLED and Motor code with it.