303 questions
- Bountied 0
- Unanswered
- Frequent
- Score
- Trending
- Week
- Month
- Unanswered (my tags)
0
votes
0
answers
60
views
How to change model body pose in an SMPL-X model
I'm quite new to modelling and 3d digital avatars. I aim to take a user video and generate a 3d model from it. I've started off simple with understanding the basics of SMPL-X, and so with the ...
Advice
0
votes
0
replies
35
views
HRNet trained from scratch on MPII high PCKh but poor visual predictions
I'm training an HRNet model from scratch on the MPII dataset for human pose estimation. The model succeded to get high accuracy around 0.93 [email protected] however the visual results are not really promising....
1
vote
0
answers
32
views
Why does the same YOLOv8n-pose model with different weights have significantly different inference speeds?
I’m testing YOLOv8n-pose models that share the exact same architecture, input size, hardware (GPU), framework, batch size, and precision settings. The only difference between them is the trained ...
1
vote
0
answers
37
views
Behavior of camera-intrinsics when downscaling image. Use-case: Megapose Pose-Estimation of industrial Objects
I'm currently working on a full pipeline to estimate the pose of industrial metal objects on a conveyor using Megapose. For that I'm using a camera with a resolution of 1280x1024px. The pipeline ...
1
vote
0
answers
102
views
Why is one of the projected points returned by cv2.projectPoints completely wrong?
I'm using cv2.solvePnP for a project. To check if the returned rotation and translation vectors are somewhat accurate, I call cv2.projectPoints on object points, but every projected point except one ...
1
vote
0
answers
71
views
GTSAM Optimizer ruins map when loop closure BetweenFactorPose3 is added
I am working on a project to use the iPhones lidar sensor for SLAM. I am trying to get GTSAM working for loop closure and I am somehow messing up the factors or frame transformations.
First I take in ...
0
votes
1
answer
68
views
How can I save a TFLite model with a fixed input shape after using interpreter.resize_input_tensor?
I am trying to run inference using the TFLite model of multipose-lightning-tflite-float16 distributed by movenet.
https://www.kaggle.com/models/google/movenet/tfLite
However, this model cannot be used ...
0
votes
0
answers
287
views
How do I do an hand-in-eye calibration with cv2.calibrateHandEye()?
I am trying to do a hand-in-eye calibration.
OpenCV Hand-Eye Calibration
In my case, I am looking for the relative transformation between a MoCap point mounted on my robot arm, usually this would be ...
2
votes
0
answers
100
views
unable to get correct camera pose from opencv Pnp algorithm
I am trying to use pnp algorithm to compute the pose of a camera which has a low fov (15 degree). The calibrated intrinsic parameters(focal length and center location) match the expected values fairly ...
0
votes
1
answer
211
views
Understanding why solvePnPRansac could retrieve bad pose estimate and how to correct it
I am trying to get pose estimation of the front face of the box (image below) using only the minimum number of points (i.e 4) by using opencv cv2.solvePnPRansac.
Here are my inputs:
The 2D pixel ...
0
votes
0
answers
111
views
How to overlay a 3D model on a physical object using pose estimation and tracking (AR application)?
I’m building an AR application where I need to estimate the pose of a physical object and overlay a 3D model of the same object on top of it, with tracking.
I’m considering using SIFT (Scale-Invariant ...
0
votes
2
answers
283
views
Size Error on YOLOv8 label format for pose estimation
Here is a sample label file:
0 0.643712 0.548266 0.0960109 0.057663 0.595707 0.519435 2 0.646533 0.526217 2 0.682304 0.548261 2 0.691717 0.577098 2
5 0.547212 0.54338 0.0545761 0.0753696 0.519924 0....
0
votes
1
answer
185
views
Yolo pose estimation output processing in Android Tensorflow Lite API
Im trying to implement yolo11n-pose_float16.tflite model to Android Kotlin project. But I cant get correct keypoints on output because there is something I did wrong. I want to draw only the skeleton ...
0
votes
1
answer
78
views
Creating different synthetic Body Shapes for one specific Human Pose for Training Pose classification
I'm working on my master's thesis, which involves tracking singing poses to provide feedback on posture. Specifically, I am using pose images, such as those derived from keypoint detection models like ...
1
vote
0
answers
271
views
Replace functionality of flutter_helper package (discontinued) to use tflite model in flutter code
tflite_flutter_helper package has been discontinued, which had provided functions to facilitate pre-processing of camera image before feeding it into the tflite model to run pose estimations. My code ...