Table 1. Recent research regarding recognition/identification/re-identification

Research areas	Reference	Target animal	Dataset	Pre-trained/transfer learning status	Feature	Algorithm
Wildlife recognition	[26]	Wildlife	Wildlife Spotter	×	−	Lite AlexNet, VGG-16, ResNet50
	[27]	Wildlife	Fishmarket, MS COCO 2017	×	−	WildARe-YOLO
Wildlife face recognition	[29]	Chimpanzee	Self-created dataset	×	Annotation Automation Framework	SSD, CNN
	[25]	Giant panda	Self-created dataset, ImageNet	○	Pre-trained AlexNet, GoogLeNet, ResNet-50, VGG-16	NIPALS
	[21]	Panda	Self-created dataset, COCO	○	Pre-trained Faster R-CNN, fine-tuned ResNet-50	DNN
	[39]	Golden snub-nosed monkey	Self-created dataset	×	−	Faster-RCNN
Livestock face recognition	[24]	Pig	Self-created dataset	×	Automatic selection of training and testing data	Haar cascade, Deep CNN
	[31]	Sheep		×	−	YOLOv5s, RepVGG
	[28]	Aberdeen-Angus cow	Self-created dataset	○	Pre-trained VGGFACE, VGGFACE2	−
	[34]	Cattle	Self-created dataset	x	Embedded system,automatically processing datasets	CNN
	[35]	Cattle	Self-created dataset	x	channel pruning	YOLOv5
identification	[37]	Cattle		×	−	Inception-V3 CNN, LSTM
	[42]	Cattle	ImageNet, COCO	x	Mobile devices	YOLOv5, ResNet18 Landmark
	[44]	Horse, etc.	THDD dataset	○	Hybrid	YOLOv7, SIFT, FLANN
Re-identification	[40]	Amur tiger	ATRW, ImageNet	○	Pre-trained SSD-MobileNet-v1, SSD-MobileNet-v2, DarkNet	YOLOv3

YOLO, You Only Look Once; MS COCO, Microsoft common objects in context; SSD, single shot multibox detector; CNN, convolutional neural network; DNN, deep neural network.