Questions tagged [computer-vision]

The goal of computer vision is an automatic extraction of information out of images. It includes methods for acquiring, processing, analyzing, and understanding images and high-dimensional data from the real world in order to produce numerical or symbolic information, e.g., in the forms of decisions. The image data can take many forms, such as video sequences, views from multiple cameras, or multi-dimensional data from a medical scanner.

Filter by
Sorted by
Tagged with
1
vote
1answer
47 views

Efficient sphere surface area search

As shown in the figure, there are objects of arbitrary shapes placed on the surface of the sphere at arbitrary locations. The problem is to search the surface to locate these objects in most efficient ...
0
votes
0answers
14 views

Hunt for fast plane extraction algorithms from sparse map points

My input was a set of sparse 3D map points(landmarks) maintained by SLAM algorithm, z-axis of the world frame permanently points to the negative direction of the ...
1
vote
2answers
24 views

How does structured light improve accuracy in stereo vision?

In my understanding, in order to improve accuracy for the estimation of depth with a stereo vision camera, patterns of light are projected into the scene, because the resulting texture makes it easier ...
1
vote
1answer
36 views

What is the relation between FOV, Focal Length and Sensor size in a Pin hole Camera Model?

I aim to recreate the camera parameters of my camera in the Unity rendering system, which has the following variables: 1) I tweaked the focal length and saw that the Field of View is changing (...
0
votes
0answers
6 views

Remove jittering from head pose estimated via dlib face detector

I am using dlib’s face detector with openCV to measure head pose and then use these values to render a Vr scene in unity 3D. The only problem is the jittering in the values estimated via computer ...
2
votes
0answers
16 views

Is it possible to calculate a homography matrix for any 2 views?

I've been experimenting with opencv to estimate the homography matrix between 2 images captured with my computer's webcam. I have noticed that when the 2 views are not very far apart (that's when I ...
2
votes
0answers
14 views

Undo pitching effect of camera when pose (y-axis) is near 90 degrees?

I am using KITTI odometry dataset whereby the camera's axis is pointing X (east), Y (down) and Z (north). It also contain the frame to frame poses of the camera (first pose is fixed as the map frame). ...
0
votes
1answer
48 views

vision guided robotic arm always required 6d pose of object? [closed]

I have an object like attached in image. It is hanging vertical (its position may change but static object) and I want to grasp it using 6DOF Robotic arm. I am Using camera to detect object. I am ...
2
votes
1answer
31 views

Short range stereo vision

I am designing a stereo vision system for short depth range (15 cm to 35 cm) and I have some problems about baseline, focal length and depth range. My application is to scan an animal tissue (meat, ...
0
votes
1answer
28 views

For pose estimation, what is the typical range of the rotation values that are estimated? Is it from -180 to 180 degrees?

I am currently working with the Rutgers APC 2015 data set for pose estimation, and see that the pose estimates, specifically the rotation estimates, have seemingly been rescaled to range between -1 ...
0
votes
1answer
25 views

Obstacle width measurement

On non-lidar equipped rovers, how is the width of an obstacle detected such that it can be avoided? If the ultrasonic sensors are used for obstacle detection, these sensors usually don't give data ...
1
vote
0answers
69 views

robot Hand Eye Calibration problem

I am trying to understand the implementations and the relative performance of various classical and new Hand-eye calibration approaches. Note: The following discussion assumes implementations are in ...
1
vote
1answer
33 views

How to perform active search in point feature based monoSLAM?

I am modifying an implementation of SLAM with single camera, MonoSLAM [1]. Instead of image patches, I want to use features points (ORB) to track landmarks. MonoSLAM uses a EKF framework. So for ...
1
vote
0answers
44 views

Tracking object in realtime using depth image stream from a RGBD camera

I have an Intel RealSense camera which can stream RGB and depth image. Using those information I want to track an object moving on a table top. More specifically, I want to know position $(x,y)$ and ...
1
vote
3answers
149 views

Why do robots need rangefinders while animals don't?

Will vision based system eventually prevail rangefinders based systems given that the most successful autonomous agents in the world like animals use primarily visions to navigate?
0
votes
1answer
54 views

Transformation matrix between a 2D and a 3D coordinate system

I have 2 sensors : one has 2D coordinate system and the other has a 3D coordinate system (not same origin). Knowing the positions of some corresponding target points (used as fixed reference) in both ...
3
votes
2answers
72 views

Question about Lens distortion

Suppose I have a 240x180 camera with the following distortion co-efficients, $$k_1 = -0.138592767408$$ $$k_2=0.0933736664192$$ $$k_3=0.0$$ $$p_1=k_4=-0.000335586987532$$ $$p_2=k_5=0.000173720158228$$ ...
2
votes
1answer
85 views

Fixing the first node in Graph-based SLAM

The first node in the graph SLAM should be fixed. The famous "a tutorial on graph based slam" paper is showing that we can fix a node by adding identity matrix. Why adding identity to the Hessian of ...
1
vote
1answer
65 views

How to tringulate many projections of a point to optimal postion?

Suppose we have many images for the same scene (with known poses already,using for ex: a GPS ), and the same feature appear in more than 2 images . How to use all the available info to compute as ...
2
votes
1answer
69 views

Robotic Hand Grasp Planning - How do I find the initial contact point for each finger when grasping an object?

I am trying to design a robotic hand that is able to adaptively grasp unfamiliar objects. I am initially thinking that using OpenCV for pose estimation might make doing so easier. Assuming that I am ...
1
vote
0answers
88 views

How to update a new pose in iterative pose optimization problem?

There are multiple ways to update new pose in iterative pose optimization problem. The easiest one we often find in papers in robotics is as SO3 + translation Update on right \begin{equation} \...
0
votes
2answers
60 views

Any reference for “3D feature matching + ICP”?

One of the simple tricks in getting high performance for a global 3D point cloud registration is combining "3D feature matching + ICP". For example FPFH matching + RANSAC for initial pose estimation ...
4
votes
3answers
318 views

What is the best SE3 library for python?

I am currently googling on this topic but I would like to hear your opinion. What is the best self-contained SE3 library that offers interconversion between quaternion, rotation vector, ...
1
vote
0answers
37 views

Algorithm for orientation (pan/tilt) towards a moving point

Main Question: note: The code in this question is pseudo-code, I'm using Python, but my pseudo-code is a mix of Python and C++. I have a camera on the top of my robot's "head" which has pan and tilt ...
0
votes
1answer
33 views

Combining multiple constraints in optimization

Let's say we have two equations for a pose $\textbf{R},\textbf{t}$ optimization problem. For ICP constraint $$\textbf{e}{_{I}}=\textbf{n}^\top(\textbf{p}_d-(\textbf{R}\textbf{p}_s+\textbf{t}))$$ For ...
1
vote
1answer
139 views

Pictures coordinates to robot coordinates

I have a picture on my C# software and the corresponding object in real world on 2d plane of motorized XY axis table. What I want is when I select a pixel on my picture, the motorized table should be ...
2
votes
0answers
145 views

How does marginalization of variables effect least squares SLAM energy function?

Consider a simple example of Bundle Adjustment where I have robot and landmark poses $x = \left[ x_p \text{ } x_m\right]^T$ and measurements given by $z$, such that a simple factor graph can be ...
1
vote
1answer
48 views

How can I receive Genicam packets from a device in a docker container?

I installed a command line tool which accepts genicam packets from a device as input into a docker image. If I use the option --network host in the ...
2
votes
2answers
388 views

Why do we need a marginalization in Bundle Adjustment?

A common practice in Bundle Adjustment is to reduce the state dimension by marginalizing structure or pose states to improve the optimization speed. In case 3d points(structure) $\textbf{p}_i$ are ...
0
votes
2answers
50 views

Estimation of Focus of expansion

I am trying to estimate focus of expansion for a moving camera mounted in a mobile robot. I am using the method described in this method page 13-14. My aim is to use it to avoid obstacle using optical ...
1
vote
0answers
124 views

Which Kinect to use for robotic arm pick and place application with object detection?

I am building a robotic arm for pick and place application. The robot is going to recognize several objects using the RGB feed from Kinect (will use a model such as YOLOv2 for object detection, ...
0
votes
1answer
52 views

Calculating covariance of parameter estimation from Hessian

I need a covariance of parameter estimation for reliability estimation. According to the this post, I know the approximation of the Hessian can be calculated as $$ \sigma^2 H^{-1} = C $$ where H is ...
5
votes
1answer
94 views

Can you find object distance using camera pose

I have bounding boxes detected in RGB images, but no information about the camera's intrinsic properties. Using deep learning (preferably) can I find the distance to the object from the camera in real ...
0
votes
1answer
183 views

Rotation composition When using Kalman Filter

I am implementing a Kalman Filter for the following situation. I have a camera set in a room that can detect the position and orientation of a marker (ARUCO) in the room. Therefore I have the ...
1
vote
1answer
92 views

Perspective N point solution: Position drifting linearly with rotation angles?

I am using a PNP algorithm to compute the rotation and translation of a camera given pre-mapped 3D points, and their corresponding projections on the 2D plane. The algorithm I am using is the one ...
1
vote
1answer
161 views

Methods for odometry/IMU/Gyro free lidar pointcloud registration for pose estimation

I am currently trying to register a pointcloud in time to find my change in position and heading at each timestep (High speed application). So this is essentially an implementation of SLAM. I am ...
0
votes
1answer
114 views

How can I compute the covariance of a camera-based relative pose measurement?

I am trying to compute the relative pose between two cameras using their captured images through the usual way of feature correspondences. I use these feature matches to compute the essential matrix, ...
2
votes
0answers
99 views

Monocular SLAM depth reconstruction

I am trying to derive the inverse depth equation used in monocular SLAM algorithms. Specifically the equation 8 of the ORB SLAM paper I am very close, except I do not understand where the (u_j-c) ...
0
votes
2answers
122 views

TU Munich Monocular Visual Odometry Dataset has NaN values in the ground truth data.

I am trying to use the motion sequences in the dataset to create trajectories for VR systems. A major portion of the ground truth data in almost all sequences has NaN values. Is there any other source ...
0
votes
1answer
124 views

How can I estimate depth of a feature point / object using single camera?

I am trying to estimate depth of a feature point that i find in my image. I read this paper that estimates the depth using a motionless camera and an object of known height to calculate horizon line. ...
1
vote
2answers
107 views

Depth sensing for a pick-and-place robot

I am supposed to design a pick-and-place robot. I have a prismatic joint to go down to a particular depth to pick an object which is located in 2D using computer vision. What is the recommended way to ...
2
votes
1answer
166 views

How to build a robotics ontology?

According to some papers about robotics soccer Ontology Driven Development of Domain-Specific Languages, 2000 and Domain knowledge for surveillance applications, 2007, i need an ontology for ...
0
votes
1answer
53 views

Pointing at a moving object

I'm a complete novice when it comes to robotics and/or computer vision, so I'm looking for a place to start - simply googling has not located what I want. I want to build a device that continuously ...
0
votes
1answer
300 views

Relative scale in SfM

I'm using monocular visual odometry. As in this case absolute scale can't be recovered, we rather compute relative scale for subsequent transformations as in this tutorial To do this, triple matches ...
2
votes
1answer
3k views

Determine the relative camera pose given two RGB camera frames in openCV-python

I would like to determine the relative camera pose given two RGB camera frames. I assume there's overlap in field of view between the two cameras, what I am looking for ultimately is the rotation and ...
2
votes
2answers
108 views

Smart camera image processing on embedded GPU vs. in the cloud

I'm working on a smart camera that does key point prediction (predicting locations of wrists, elbow, shoulder, ears, eyes, nose, etc.) for gesture recognition. Right now, the neural net is running on ...
1
vote
1answer
151 views

Kalman filter for vision based pose estimation: 'good' measurements not improving system covariance

I am starting off with a very simple Kalman filter for vision based pose estimation (PnP algorithm). The filter is inspired by the constant velocity model in this OpenCV tutorial, but I am ignoring ...
2
votes
1answer
196 views

OpenCV get topdown view of planar pattern by using intrinsic and extrinsic from cameraCalibrate

Originally I have a image with a perfect circle grid, denoted as A I add some lens distortion and perspective transformation to it, and it becomes B In camera calibration, A would be my destination ...
1
vote
2answers
544 views

Scan an Auditorium, Count People, and Display Open Seats [closed]

I'm looking for any recommended approaches to be able to digitally process a room full people, like an auditorium or a movie theater and output the following information: How many people are in the ...
1
vote
1answer
809 views

How many AR markers should I use to get the object pose?

I'd like to get the object pose based on the attached AR marker to the object. I believe I only need one tag to get the 6D pose. However, I sometimes see more than one AR markers are used in the ...