Object detection (Images)

The two most common image-processing problems are image classification and object detection.

Image classification takes an image as an input and outputs what type of object is in the image. This technique works great, even on microcontrollers, as long as we only need to detect a single object in the image.

On the other hand, object detection takes an image and outputs information about the class and number of objects, position, (and, eventually, size) in the image.

Edge Impulse provides four different methods to perform object detection:

• Using MobileNetV2 SSD FPN

• Using FOMO

• Using NVIDIA TAO

• Using YOLOv5 (Custom Learn Block)

Specifications	MobileNetV2 SSD FPN	FOMO	NVIDIA TAO	YOLOv5
Labelling method	Bounding boxes	Bounding Boxes	Bounding boxes	Bounding boxes
Input size	320x320	Square (any size)	Flexible	Flexible
Image format	RGB	Greyscale & RGB	RGB	RGB
Output	Bounding boxes	Centroids	Bounding boxes	Bounding boxes
MCU	❌	✅	✅	✅
CPU/GPU	✅	✅	✅	✅
Limitations	- Works best with big objects - Models use high compute resources (in the edge computing world) - Image size is fixed	- Works best when objects have similar sizes & shapes - The size of the objects are not available - Objects should not be too close to each other	- Works best on high end MCU	- More compute intensive - Not suitable for all edge devices