Stack Overflow
Stack Overflow

Questions tagged [pose-estimation]

The specific task of determining the position and orientation of an object in an image relative to the camera coordinates is called Pose Estimation.

The pose of a camera is also known as the camera Extrinsic Parameters. It is a 3x4 matrix where the first 3 columns are the rotation (orientation) and the las column is the translation (position).

Camera Pose matrix. Rotation and Translation

291 questions
0
votes
2 answers

How to get one of the specific values(with the specific index) of this iterable enumerate (results.pose_landmarks.landmark)?

import cv2 #OpenCV is the library that we will be using for image processing import mediapipe as mp #Mediapipe is the framework that will allow us to get our pose estimation import time mpDraw = mp.solutions.drawing_utils mpPose =…
Matt095
  • 857
  • 3
  • 9
0
votes
1 answer

Tensorflow pose estimation strange behaviour

I am trying to detect body parts/landmarks from pictures but I'm facing an issue. For some reason it prints the knee points even when there's no knees in the picture. Any ideas how and why to fix this ? Or what is the better/faster way to detect…
Andie31
  • 305
  • 3
  • 13
0
votes
1 answer

pose estimation on Triton inference server

I am struggling with running pose models in NVIDIA Triton inference server. The model (open pose , alpha pose , HRNet ... etc ) load normally but the post processing is the problem
younes
  • 1
  • 1
0
votes
1 answer

Landscape orientation on Media pipe Android App.. The camera orientation is wrong once the app is set in landscape mode

I created media pipe android app. And remove the restrictions so that it can work in landscape. But the camera is rotated by 90 degrees in landscape mode. Is there any solution for this?
christy varghese
  • 1
0
votes
1 answer

RGB-Webcam for 3D pose estimation

I don't know whether this is the right place to ask this question but here we go. I'm currently implementing a project for 3D pose estimation and with VOXELPOSE. (based on OpenCV) Therefore, I want to set up at least two cameras to optimize the 3D…
pacours
  • 33
  • 6
0
votes
1 answer

How to run densepose on video with Detectron2

I was wondering if it is possible to run densepose annotations on a mp4 with detectron2? In the projects folder, you can run densepose with applynet.py but this only works on images. I tried running this commmand d demo/ python demo.py…
Dark Apostle
  • 189
  • 4
  • 18
0
votes
1 answer

"torch.relu_(input) unknown parameter type" from pytorch

I am trying to run this 3D pose estimation repo in Google Colab on a GPU, but after doing all of the steps and putting in my own left/right cam vids, I get this error in Colab: infering thread started 1 1 : cannot connect to X server Exception in…
John Adams
  • 109
  • 10
0
votes
0 answers

Transformation Matrix From EstimatePoseSingleMarkers

I try to get transformation matrix from my camera to the mid point of an aruco marker and I use cv2.aruco.estimatePoseSingleMarkers function. In the description it says : The returned transformation is the one that transforms points from each marker…
Thompsoncikl
  • 1
  • 1
0
votes
1 answer

Python library to retrieve face pose from 68 landmark points

Can anyone point me in the direction of a library that can estimate face pose from 68 landmark points in Python? To be more specific, many such libraries exist. However, they usually start from an image. I want to skip the landmark detection because…
djip.co
  • 997
  • 10
  • 17
0
votes
1 answer

Can solvePnP handle points outside of the Image-frame?

Kontext: I have a soccer field, and I am building a Pipeline for Camera calibration (far from actual implementing). The camera is static and has an overview of the field. After some preparation I want to take the crossings of the field-lines to…
Mike Heinemann
  • 1
  • 1
0
votes
1 answer

Is there a python only version of mediapipe?

I am currently working with a JeVois camera. Code I have previously written for pose estimation with high performance and only using a cpu uses Mediapipe. However, JeVois is not very friendly with added libraries like mediapipe as it is mostly c++…
Stradtdog
  • 1,490
  • 2
  • 7
  • 13
0
votes
0 answers

What's the difference b/w RANSAC-based voting & Hough voting to estimate vector-field for keypoints

The paper "PVNet: Pixel-wise Voting Network for 6DoF Pose Estimation" use RANSAC based voting for localizing keypoints, and further use PnP to calculate object poses. The paper "6DoF Object Pose Estimation via Differentiable Proxy Voting…
ML Dev
  • 69
  • 1
  • 7
0
votes
0 answers

Aruco marker : get orientation and position after pose estimation

I want to know the real position and orientation of my marker. If I understood well, tvecs gives me the real position of my marker, I measure the distance with a ruler and it seems be correct even I had sometimes weird values. If someone has an idea…
Newbi
  • 31
  • 6
0
votes
0 answers

How to use TensorFlow for comparing key points?

I am working on a fitness android application that would detect if a person has exercised correctly or not. The app will record the person exercising, then compare it with the models, and display if the person has exercised correctly or not. I have…
Maria Salim
  • 1
  • 4
0
votes
0 answers

How a CNN based network developed for 2D RGB data, be trained on 3D data like LINEMOD dataset for object pose estimation?

PoseCNN (PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes) is using a CNN as backbone network. CNNs can be trained on 2D RGB data. How can we train the PoseCNN using 3D data (RGB-D like) like LINEMOD,…
ML Dev
  • 69
  • 1
  • 7
1 2 3
…
19 20