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
1
vote
1 answer

Why are pose estimators so slow?

I'm new to the ML field and have a question about computer vision. Why can face detection and object detection software (such as Instagram's face filters) be run on our smart phones at 30 - 60 FPS smoothly, while pose estimation software (e.g…
RodYt
  • 157
  • 2
  • 4
1
vote
1 answer

Is a point inside a polygon? Problem with method

First post in stackoverflow. Im working with TfPoseEstimator in python and need to know if a point of the human body is inside an area that I delimited in a polygon. My problem is that the try block says: "Not working" for each part of the…
Luis Bleda Torres
  • 25
  • 5
1
vote
0 answers

Move an object between two ml5js poseNet keypoints using p5js

I'm trying to make a circle move between two body keypoints – i.e between left and right shoulder. I started with this code to make an object following some position: var rectLocation; var x = 50; var y = 50; function setup() { …
sebyakin
  • 11
  • 1
1
vote
1 answer

How to detect the angle between two lines in real time by tensorflow

Our graduation project is to track human motion at real time while doing an exercise and then give him feedback if that exercise is right or wrong and instructions to make it a perfect one, so we split the code into two parts. The first one is…
Aya N. Mohammed
  • 31
  • 6
1
vote
0 answers

SolvePnp doesn't give the same result with four points and more points

I'm computing the pose of the camera using SolvePnP function of OpenCv library and I'm facing the issue that when I give four point it gives me slightly better result than if I give more points. Here is an example of the code with some points as an…
Ja_cpp
  • 2,426
  • 7
  • 27
  • 49
1
vote
3 answers

Generating keypoint heatmaps in Tensorflow

I am trying to train a model for facial keypoints detection. This is a Stacked HourGlass model. It outputs 256x256x68 dimensional tensor. Each of the 68 outputs will have a hot region around a keypoint. I have defined the model and graph…
azmath
  • 863
  • 1
  • 11
  • 30
1
vote
2 answers

Correctly interpreting the Pose (Rotation and Translation) after 'recoverPose' from Essential matrix in OpenCV

I have been breaking my head trying to correctly interpret the results of recoverPose from Essential matrix. This is based on original code posted in How do I estimate positions of two cameras in OpenCV? Here are the high level steps I am using: 1.…
Vik
  • 690
  • 7
  • 22
1
vote
1 answer

Parameter "triangulatedPoints" of recoverPose and math behind it

cv::recoverPose has parameter "triangulatedPoints" as seen in documentation, though math behind it is not documented, even in sources (relevant commit on github). When I use it, I get this matrix in following form: [0.06596200907402348,…
stackoverflower
  • 545
  • 1
  • 5
  • 21
1
vote
0 answers

stability of AruCo markers

I am using 3 AruCo markers of size 1cm each for head pose estimation in C++ using OpenCV library on live webcam stream. I am able to detect the markers and their pose accurately but had issues with stability of the pose. I used the function…
adi1992
  • 75
  • 11
1
vote
1 answer

More complex hand-pose estimation algorithms

I am currently looking into hand-pose estimation in Unity without using any expensive plugins! At the moment, I have implemented a simple hand-tracking system by extracting the contours of the hand, like the link…
Samuel Lopes
  • 35
  • 8
1
vote
1 answer

Averaging SIFT features to do pose estimation

I have created a point cloud of an irregular (non-planar) complex object using SfM. Each one of those 3D points was viewed in more than one image, so it has multiple (SIFT) features associated with it. Now, I want to solve for the pose of this…
John B
  • 159
  • 1
  • 5
1
vote
0 answers

Determine angle of rotation for a skewed 2D image

I am trying to determine the angle of rotation required to fix the skew seen in a 2D image. I have a camera that is looking at a QR code taped to a wall. I am trying to find out how the camera should rotate left and right so that the camera is…
Rachel
  • 91
  • 2
  • 11
1
vote
2 answers

Plotting the normal vector of a plane in OpenCV

I'm using a 2D barcode to identify a plane in 3D space and I want to plot its normal with respect to its center. This is the code I'm using to compute the normal def compute_normal(camera, board, bounds, frame): extrinsics, ip, op =…
Carpetfizz
  • 8,707
  • 22
  • 85
  • 146
1
vote
1 answer

How to estimate face pose from 5 face landmarks in dlib ?

I saw recently the 5 point landmarking model from dlib: https://github.com/davisking/dlib-models I am instereing to use it to pose estimation, but I do not find the function in dlib that perform it. My question does it exist in dlib (C++,…
Viktoriia
  • 85
  • 1
  • 2
  • 10
1
vote
0 answers

OpenCV SolvePNP returns the same results in different runs

As far as I know solve pnp algorithm (almost all of the pose estimation algorithm - or at least all cv.solepnp flags) suppose to produce a different answer in every run because of the idea of estimation in the first place. I'm trying to check the…
Guy P
  • 1,395
  • 17
  • 33
1 2 3
…
19 20