Reduced model performance when tried with another face detection model

[brittosabu]

• face_recognition version:
• Python version:3.7.4
• Operating System: Ubuntu 10.04

Description

The detection model was not performing as per our needs , so we tried Yolov3 model for face detection. The face detection works perfectly now, but the recognition is very bad and it keeps changing the 'Names' for the identified people. We've tried changing tolerance, and almost everything, is there something which we are missing out

[haris-ahmed]

try following for face detection https://www.pyimagesearch.com/2018/02/26/face-detection-with-opencv-and-deep-learning/

when you pass an image to Yolov3 it tries to detect following in supplied image:

person, bicycle, car, motorbike, aeroplane, bus, train, truck, boat, traffic light, fire hydrant, stop sign, parking meter, bench, bird, cat, dog, horse, sheep, cow, elephant, bear, zebra, giraffe, backpack, umbrella, handbag, tie, suitcase, frisbee, skis, snowboard, sports ball, kite, baseball bat, baseball glove, skateboard, surfboard, tennis racket, bottle, wine glass, cup, fork, knife, spoon, bowl, banana, apple, sandwich, orange, broccoli, carrot, hot dog, pizza, donut, cake, chair, sofa, pottedplant, bed, diningtable, toilet, tvmonitor, laptop, mouse, remote, keyboard, cell phone, microwave, oven, toaster, sink, refrigerator, book, clock, vase, scissors, teddy bear, hair drier, toothbrush

it detects person not only face I don't think using Yolov3 is good idea for face detection.

[brittosabu]

@haris-ahmed I've used a Yolo detector specifically trained to detect faces only. So it gets almost every face in the image. The problem is with the recognition part

[haris-ahmed]

can you share code or steps you are doing for recognition part and how many trained faces do you have for a person?

[448357739]

Hi,I have the same question when processing the video. Do you have any ideas to solve it now?