FSA-Net
FSA-Net copied to clipboard
[CVPR19] FSA-Net: Learning Fine-Grained Structure Aggregation for Head Pose Estimation from a Single Image
Bumps [pyyaml](https://github.com/yaml/pyyaml) from 5.1.2 to 5.4. Changelog Sourced from pyyaml's changelog. 5.4 (2021-01-19) yaml/pyyaml#407 -- Build modernization, remove distutils, fix metadata, build wheels, CI to GHA yaml/pyyaml#472 -- Fix for...
Hi, Thank you very much for providing the code! I am trying to reproduce the results provided in the paper for protocol 1. I used the processed dataset from the...
Hi, I'm trying to convert your pretrained frozen graph to TF-TRT using this code: ```python import tensorflow as tf from tensorflow.python.compiler.tensorrt import trt_convert as trt with tf.Session() as sess: #...
Now, I 'm trying to convert model to TFlite. But it has some problems with custom layers. Did anyone convert it successfully?
Did anybody able to convert these models into onnx format? @shamangary @ksachdeva
I make a better one [https://github.com/WIKI2020/FacePose_pytorch]
I met a problem that the bounding box laying outside of my input image (360x640). Something like this. [https://i.imgur.com/Tdcu0bU.png](https://i.imgur.com/Tdcu0bU.png) So when i run into this line `faces[i,:,:,:] = cv2.resize(input_img[yw1:yw2 +...
Instructions for updating: dim is deprecated, use axis instead WARNING:tensorflow:From e:\sundwpyworkspace\vscode\FSA-Net\demo\lib\capsulelayers.py:195: calling softmax (from tensorflow.python.ops.nn_ops) with dim is deprecated and will be removed in a future version. Instructions for updating:...
Hi team, Just curious what was the final loss after training a model? For me, it was around 12 MAE. is that normal? Thanks