Questions tagged [object-recognition]

The computer vision task of finding and identifying objects in an image or video sequence

The computer vision task of finding and identifying objects in an image or video sequence is often conducted consistent with one of two approaches:

  • Appearance-based methods

    • Use example images (called templates or exemplars) of the objects to perform recognition

    • Objects look different under varying conditions:

  • Feature-based method

    • a search is used to find feasible matches between object features and image features.

    • the primary constraint is that a single position of the object must account for all of the feasible matches.

http://en.wikipedia.org/wiki/Outline_of_object_recognition

316 questions
1
vote
0 answers

EmguCV detect keypoints from 3D Model

Is there a way i can retrieve the observed keypoints from a 3D model instead from an image. This is since i need to track an uneven object (simple spaceship right now: http://i.msdn.microsoft.com/dynimg/IC129855.jpg) that can be visible from any…
Pawcu
  • 332
  • 1
  • 14
1
vote
3 answers

How to detect groups of circles with openCV?

I have a picture like that below. I would like to find groups of circles (their positions) in the image. In the following example there should be three groups. The background is white or will be whitish color. (In the source image there will not…
John
  • 1,834
  • 5
  • 32
  • 60
0
votes
0 answers

What is going wrong when trying to apply a gabor filter to an image?

I am trying to use a gabor filter on an image using a gabor function and parameters from a published object recognition paper. Instead of outputting the filtered image, the output image seems to be a portion of a white/gray circular gabor gradient…
0
votes
0 answers

Matching unique objects in a collection of images using object detection and similarity measures

I have a collection of photographs depicting various objects taken from a very large dataset. Each image contains a different number of objects, and the same object can be depicted slightly differently. I am able to detect the objects in the images…
zlon
  • 812
  • 8
  • 24
0
votes
0 answers

How do I operate the object recognition code?

I'm using a model for detecting objects on my Raspberry Pi, using python. This is the python code: import numpy as np import cv2 import os,time import tensorflow as tf from object_detection.utils import label_map_util from object_detection.utils…
0
votes
1 answer

YOLO multiple annotations on same object

Is it ok to put multiple annotations of the same class on the same object? Is it going to break or not affect or increase accuracy? OneBirdMultipleAnnotations Iv put multiple bounding boxes of the same class on an object
0
votes
0 answers

How to draw squares around objects in image and count them?

So, I made a (hopefully good) model in python for apples; I used a dataset from Kaggle that had a bunch of pictures of apples upclose, and trained it to recognize them with accuracy of around 85%. My question now is how do I use that model to draw…
0
votes
2 answers

Images labeling for object detection when object is larger than the image

how I should label objects to detect them, if the object is larger than the image, e.g. I want to label a building, but in the picture is visible only part of the building (windows and doors, without roof). Or should I remove these pictures from my…
0
votes
1 answer

Is there a right way to programmatically prevent a brief wrong recognition (in object detection app) to trigger an action?

Context I'm building an app which performs real-time object detection throught the camera module of the device. The render is like the image below. Let's say I try to recognize an apple, most of the time the app will recognize an apple. However,…
0
votes
0 answers

Detect objects from live YouTube video and generate links

I am working on a small project where I'd like to create a browser extension to work with Youtube (or other streaming sites) to detect trendy items in the video and generate purchase links which will be shown in a column (the idea is not completely…
Satashree Roy
  • 365
  • 2
  • 9
0
votes
0 answers

How to draw straight line between several frames with moving camera?

I am trying to get ball coordinates data from football (soccer) video. From technology that I have used I was able to determine ball coordinates. Example, 12 second video was divided into several frames and from each frame I should have got ball…
0
votes
2 answers

General object recognition with biggest number of classes

I'm new to the computer vision world, I'm trying to create a script with the objective to gather data from a dataset of images. I'm interested in what kind of objects are in those images and getting a summary of them in a json file for every…
0
votes
1 answer

How do I trigger a JSON or pd Dataframe report from each Object detected by a TensorFlow/OpenCV object detection

I'm building a python application that uses AI to scan a live feed with OpenCV and tracks people with or without masks. I wanted to ask if there was a good way to send a unique dataframe for every unique entity tracked in the feed in order to create…
0
votes
1 answer

Google Cloud Function - Video Intelligence

I know how to trigger a video intelligence request for object tracking / object detection with Google Cloud as following : video_client = videointelligence.VideoIntelligenceServiceClient() features =…
0
votes
1 answer

How to draw a polygon shaped data annotation shape for image detection?

I am training some images using imageAI object detection. What i want to do is draw a polygon shaped data annotation around the objects in an image. Not a square box. Even though I ca draw the shape, the imageAI class doesn't consider polygon…