#cplusplus #caffe #computer_vision #deep_learning #digits #embedded #image_recognition #inference #jetson #jetson_agx_xavier #jetson_nano #jetson_tx1 #jetson_tx2 #jetson_xavier #machine_learning #nvidia #object_detection #robotics #segmentation #tensorrt #video_analytics

https://github.com/dusty-nv/jetson-inference

#python #darknet2onnx #darknet2pytorch #onnx #pytorch #pytorch_yolov4 #tensorrt #yolov4

https://github.com/Tianxiaomo/pytorch-YOLOv4

#cplusplus #vgg #resnet #alexnet #squeezenet #inceptionv3 #googlenet #resnext #tensorrt #crnn #senet #arcface #mobilenetv2 #yolov3 #shufflenetv2 #mnasnet #retinaface #mobilenetv3 #yolov3_spp #yolov4 #yolov5

https://github.com/wang-xinyu/tensorrtx

#python #tensorrt #ncnn #onnx #yolov3 #openvino #yolox #yolox_nano

https://github.com/Megvii-BaseDetection/YOLOX

#cplusplus #deployment #model_converter #ncnn #onnxruntime #openvino #pplnn #sdk #tensorrt

https://github.com/open-mmlab/mmdeploy

#python #detection #detextron2 #detr #face #instance_segmentation #object_detection #onnx #tensorrt #transformers #yolo #yolov6 #yolov7 #yolox

https://github.com/jinfagang/yolov7

#cplusplus #android #deep_learning #deployment #graphcore #intel #ios #jetson #kunlun #object_detection #onnxruntime #openvino #picodet #rockchip #sdk #serving #tensorrt #uie #yolov5

https://github.com/PaddlePaddle/FastDeploy

#python #anchor_free #computer_vision #deep_learning #edge_computing #object_detection #object_detector #onnx #pytorch #tensorrt #yolo

https://github.com/LSH9832/edgeyolo

#python #damo_yolo #deep_learning #imagenet #nas #object_detection #onnx #pytorch #tensorrt #yolo #yolov5

https://github.com/tinyvision/DAMO-YOLO

#python #botsort #bytetrack #deep_learning #deepocsort #improvedassociation #mot #mots #multi_object_tracking #multi_object_tracking_segmentation #ocsort #osnet #segmentation #strongsort #tensorrt #tracking_by_detection #yolo

BoxMOT is a tool that helps track multiple objects in videos or images using advanced models. It offers various tracking methods that work well on different types of hardware, from CPUs to powerful GPUs. This means you can use it even if your computer is not very powerful. BoxMOT also saves time by allowing you to reuse pre-generated data, so you don't have to repeat calculations every time. You can easily install and use it with popular object detection models like YOLOv8, YOLOv9, and YOLOv10, and it supports tracking different types of data such as bounding boxes, segmentation masks, and pose estimations. This makes it very flexible and useful for various tasks involving object tracking.

https://github.com/mikel-brostrom/boxmot