PoseEstimationForMobile

PoseEstimationForMobile copied to clipboard

Bumps [numpy](https://github.com/numpy/numpy) from 1.13.3 to 1.22.0. Release notes Sourced from numpy's releases. v1.22.0 NumPy 1.22.0 Release Notes NumPy 1.22.0 is a big release featuring the work of 153 contributors spread...

dependabot[bot]

Bumps [tensorflow-gpu](https://github.com/tensorflow/tensorflow) from 1.4.0 to 2.7.2. Release notes Sourced from tensorflow-gpu's releases. TensorFlow 2.7.2 Release 2.7.2 This releases introduces several vulnerability fixes: Fixes a code injection in saved_model_cli (CVE-2022-29216) Fixes...

dependabot[bot]

You mentioned "I transfer the annotation into COCO format for using the data augument code from tf-pose-estimation repository." However, the repo "tf-pose-estimation" doesn’t exist. Would you please address this? Actually,...

AzinPoshtyar

Model size - I trained a model by the repo's code and recieved model with bigger size than the release model

1

Hello I take your original github implementation, with the `mv2_cpm.cfg` config file and the `network_mv2_cpm.py`. I trained a model for 24K steps and produced `.pb` model file. Your `.pb` file...

yonatanbitton

uint8 quantization

1

Hello! Has anyone tried uint8 quantization of this model? I tried to use this script: tflite_convert --graph_def_file=./model.pb --output_file=./pose-quant.tflite --input_format=TENSORFLOW_GRAPHDEF --output_format=TFLITE --inference_type=QUANTIZED_UINT8 --input_shape="1,96,96,3" --input_array=image --output_array=Convolutional_Pose_Machine/stage_5_out --default_ranges_min=0 --default_ranges_max=6 --mean=0 --std_dev=1 After that...

Jove125

我想在google colab上训练这部分数据，有指导吗，我看了要不就是docker要不就是mac

1

dandingol03

resume from pretrained

1

Hi, is there an option to resume training from checkpoint?

dodler

Using my own dataset how to define the keypoints in json file?

1

Hi, I got this running and training with the ai_challenger data and now I'd like to try some of my own data. Bu I don't quite understand the whole format...

esteimle

训练结果模型转换为MACE模型问题

8

使用PoseEstimation训练框架训练出模型，并通过以下代码段转换为frozen-pb模型； `# Convert to frozen pb. cd training python3 src/gen_frozen_pb.py \ --checkpoint=/model-xxx --output_graph=/model-xxx.pb \ --size=192 --model=mv2_cpm_2` 然而，将frozen-pb模型转换为MACE模型后，并通过以下代码段进行MACE模型校验： `python tools/converter.py run --config=mace_new/builds/downloads/test.yml --validate` 校验时，报warning如下： `Generate input file: builds/testSrcFrozen/_tmp/frozen_src/99da42816155dfdcf80ef3ba3a67028b/general/arm64-v8a/model_input_image Generate input file...

lvchigo

Hi all the size of PoseEstimationForMobile/release/cpm_model/model.pb is 2666kb. While , I convert your model(PoseEstimationForMobile/blob/master/training/src/network_mv2_cpm.py) to pb file by myself.The size of my model is 5237kb. what's more, the speed of...

PuAnysh