martlgap / faceidlight

A lightweight face-recognition toolbox and pipeline based on tensorflow-lite

Hi there 👋 it's Martin,

I am research associate at the Institute of Human-Machine Communication, Technical University of Munich, with the aim of obtaining a doctorate (PhD).

Working on tasks related to cross-resolution face verification.

---

🛠 Technologies & Tools

---

📈 Stats

---

🧑🏼‍💻 Languages

---

✉️ Contact

faceidlight's Issues

Acurracy is pretty low

face bounding detections are inaccurate and cannot able to extract features from some images, and why the models are downloading everytime

try to fix this

issue with customized gallery

Hi, when I was trying to run your demo with
from FaceIDLight.demo import Demonstrator
Demonstrator(gal_dir=).run()

this error would show up:
Python 3.8.18 (default, Sep 11 2023, 08:17:33)
[Clang 14.0.6 ] :: Anaconda, Inc. on darwin
Type "help", "copyright", "credits" or "license" for more information.

from FaceIDLight.demo import Demonstrator
Demonstrator(gal_dir='./proj1_gallary').run()
Traceback (most recent call last):
File "", line 1, in
File "/opt/anaconda3/envs/ee4208_proj1/lib/python3.8/site-packages/FaceIDLight/demo.py", line 38, in init
self.FaceID = FaceID(gal_dir=gal_dir)
File "/opt/anaconda3/envs/ee4208_proj1/lib/python3.8/site-packages/FaceIDLight/tools.py", line 62, in init
self.update_gallery()
File "/opt/anaconda3/envs/ee4208_proj1/lib/python3.8/site-packages/FaceIDLight/tools.py", line 81, in update_gallery
self.gal_embs = self.recognizer.get_emb(np.asarray(self.gal_faces))[0]
File "/opt/anaconda3/envs/ee4208_proj1/lib/python3.8/site-packages/FaceIDLight/tools.py", line 144, in get_emb
return tflite_inference(self.face_recognizer, img)
File "/opt/anaconda3/envs/ee4208_proj1/lib/python3.8/site-packages/FaceIDLight/tools.py", line 120, in tflite_inference
model.allocate_tensors()
File "/opt/anaconda3/envs/ee4208_proj1/lib/python3.8/site-packages/tflite_runtime/interpreter.py", line 259, in allocate_tensors
return self._interpreter.AllocateTensors()
RuntimeError: external/org_tensorflow/tensorflow/lite/kernels/pad.cc:79 SizeOfDimension(op_context->paddings, 0) != op_context->dims (4 != 1)Node number 0 (PAD) failed to prepare.

Can I ask how to address this issue? I've checked the path to the galley folder is correct.

How to use the ResNet50 onnx models instead of tflite models

Please provide support for python 3.9