mmpose

mmpose copied to clipboard

add chart alongside to the inference speed table

1

I've plotted the https://mmpose.readthedocs.io/en/latest/inference_speed_summary.html to have a meaningful insight into the data. ```python import pandas as pd import plotly.express as px df = pd.read_html('https://mmpose.readthedocs.io/en/latest/inference_speed_summary.html')[0] df[['GPU FPS', 'GPU FPS stderr']] =...

smidm

Clarification on min_keypoints in tracking

2

Could someone explain the motivation behind this seemingly arbitrary line of code? https://github.com/open-mmlab/mmpose/blob/c8e91ff456d82c7f985bf938cabfb68b2aa51d27/mmpose/apis/inference_tracking.py#L227 Here is a copy of the relevant section with the line marked: ``` if track_id == -1:...

pallgeuer

Custom Object Keypoint detector

3

I am looking to implement a keypoint detetctor with a custom dataset. The setup for hand or full body keypoint detection or retraining is understanadble from the documentation. Is there...

nath-partha

How to get h36m_coco_test.json and h36m_coco_train.json?

3

We want to train a topdown model for h36m dataset. However, h36m_coco_test.json and h36m_coco_train.json in the /configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/h36m/hrnet_w32_h36m_256x256.py are missing.

qingshi9974

installation error via opencv-python

1

Following your installation guide, I installed mmpose and other stuff. When opencv-python==4.6.0.66 is installed, mmpose and mmcv will raise an error like this: ``` AttributeError: partially initialized module 'cv2' has...

Yamameeee

change maxDet on evaluation

1

I think some of my test images may have multiple object bboxes more than 20, so maybe the maxDet=20 in evaluation process may make mAP lower. Where can I change...

xypu98

Pretrained smoothnet model for 2D pose?

3

Can you provide a smoothnet config / pre-trained model for 2D pose inference? The existing configs are for 3D pose (H36M) instead of 2D pose (COCO)

changt

onnx infer speed is too slow

5

i used pytorch2onnx.py to transfer res50_coco_256x192.pth to onnx. also i transferd it to torchscript. i tested onnx and torchscript model infer dynamic_batch inputs. all were used gpu with cuda, onnx...

liuxufenfeiya

Roadmap of MMPose

76

We keep this issue open to collect feature requests from users and hear your voice. Our monthly release plan is also available here. You can either: 1. Suggest a new...

hellock

## Motivation Support LSTM Pose Machine. ## Modification ## BC-breaking (Optional) ## Use cases (Optional) ## Checklist **Before PR**: - [ ] I have read and followed the workflow indicated...

luminxu