Questions tagged [object-recognition]

For questions related to object recognition, which is the problem of determining the type/class/category of an object in the image, so object recognition could also be called object classification. This is different from object detection, which is either used to refer to object localization (i.e. find the coordinates of the object in the image) + object classification, or just object localization.

40 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
9
votes
1answer
254 views

Extending FaceNet’s triplet loss to object recognition

FaceNet uses a novel loss metric (triplet loss) to train a model to output embeddings (128-D from the paper), such that any two faces of the same identity will have a small Euclidean distance, and ...
5
votes
0answers
300 views

What are the ways to calculate the error rate of a deep Convolutional Neural Network, when the network produces different results using the same data?

I am new to the object recognition community. Here I am asking about the broadly accepted ways to calculate the error rate of a deep CNN when the network produces different results using the same data....
3
votes
0answers
35 views

How are Ground truth provided to each Pyramid map in RetinaNet or YOLOv3 Paper? How is the mapping of Feature Pyramids done to Ground Truth

SO the YOLO V3 and RetinaNet both uses the Feature pyramids which look something like this: (except b and e which have one ...
3
votes
0answers
41 views

Image classification - Need method to classify "unknown" objects as "trash" (3D objects)

We have an image classifier that was built using CNN with faster R-CNN and Yolov5. It is designated to run on 3D objects. All of those objects have similar "features" structure, but the ...
3
votes
0answers
29 views

If random rotations are included in the data augmentation process, how are the new bounding boxes calculated?

When studying bounding box-based detectors, it's not clear to me if data augmentation includes adding random rotations. If random rotations are added, how is the new bounding box calculated?
3
votes
0answers
42 views

Video recognition (specifically video, not individual frames)

There are libraries for recognizing individual video frames, but I need to recognize an object in motion. I can recognize a person in every single frame, but I need to know if the person is running or ...
3
votes
4answers
389 views

Can bounding boxes further improve the performance of a CNN classifier?

Suppose I have a standard image classification problem (i.e. CNN is shown a single image and predicts a single classification for it). If I were to use bounding boxes to surround the target image (i.e....
3
votes
2answers
564 views

Alternative to sliding window neural network (was: Object detect (or) image classification at specific locations in the frame)

Recent advances in Deeplearning and dedicated hardware has made it possible to detect images with a much better accuracy than ever. Neural networks are the gold standard for computer vision ...
3
votes
0answers
610 views

Getting worse performance when training a pre-trained model with the existing class

I am training pre-trained SSD-InceptionV2-Coco to detect the "car", which is one of the classes in mscoco label. I train the model with ~50k sample from KITTI, 500k iteration with batch size 2. I ...
2
votes
0answers
36 views

Is it ok to perform transfer learning with a base model for face recognition to perform one-shot learning for object classification?

I am trying to create a model that is using a one-shot learning approach for a classification task. We do this because we do not have a lot of data and it also seems like a good way to learn this ...
2
votes
0answers
32 views

Detect object in video and augment another video on top of it

I'm trying to detect an object in a video (with slight camera movement), and then augment another video on top of it. What is the simplest approach to do that? For instance, let's assume I have this ...
2
votes
0answers
37 views

YOLO 9000 about Better Stronger

In this paper, YOLO has three features compared to YOLO v1. This question is about Better and Faster. In the Better section, there are many techniques such as Batch Norm, Anchor Box and so on. In the ...
2
votes
0answers
45 views

FasterRCNN's RPN network training

I would like to know if my understanding of RPN training is correct, and if never training the RPN on some specific anchor box is bad (i.e if the anchor never sees good nor bad examples). To make my ...
2
votes
0answers
20 views

Can Microsoft's cognitive service find similar person in a set of images without using the face service?

I need to create an application that can detect if a person X entered as an input exists in an image set and return as output all the images in which the person X exists. The problem is that the ...
2
votes
0answers
49 views

How should I build an AI that quickly detects falling game assets on screen?

I want to build an AI that plays a simple android game. The game is just a one at a time object falling, some times at an angle. The AI needs to recognize the object and to decide whether to swipe ...
2
votes
0answers
62 views

Object size identification and maximum number of classes with convolutional neural networks

I am working on a project that involves using a ConvNet to identify screws. I am able to train from scratch a ConvNet based on the first version of the inception network, but shallower (only 3 ...
2
votes
0answers
27 views

Sails size recognition

Is it possible to recognize the height and width of the sails of different kitesurfers and windsurfers taken from public webcams? And show these information on video in real time? Or on screenshots?
2
votes
0answers
36 views

Presence of object (highly occluded vehicle) in a scene

How to detect presence of object (highly occluded) in a scene? There are specific features (small patterns, etc), which allow to say that object is present, but it is not enough for detection for ...
2
votes
2answers
366 views

Face liveness detection using face landmark points

How to detect liveness of face using face landmark points? I am getting face landmarks from android camera frames. And I want to detect liveness using these landmark points. How to tell if a human ...
2
votes
0answers
45 views

How a game playing agent could identify potential objects and proximity?

Most implementations I'm seeing for playing games like Atari (usually similar to DeepMind's work using DQN) have 4 graphical frames of input fed into 3 convolutional layers which are then fed into a ...
1
vote
0answers
23 views

In anchor based object detection, why don't the anchors share the same weights?

After reading about YOLO V3 and Faster R-CNN, I don't understand why the weights for the regression head aren't the same across all boxes of the same size. Given that the backbone of these systems is ...
1
vote
0answers
19 views

How can I train a model to recognize object with zoomed-in image?

Humans are good at guessing animals with zoomed-in images from patterns of fur/skin. (For example, if we saw a black-white pattern fur, it must be a zebra) I have some experience guessing a car model ...
1
vote
0answers
56 views

Can object detection approaches be used to solve text/detection problems?

I have been working on text detection and recognition for almost two months and new on this field. So far, I have fine-tuned, tested, and trained several text detection/recognition methods, such as ...
1
vote
0answers
21 views

Find object's location in an area using computer vision

I'm trying to see how to detect the location of a soccer ball in the field using the live camera. What are some ways to achieve this? 1- Assuming we have a fixed camera with a wide shot. How to find ...
1
vote
0answers
530 views

Can I resize my images after labeling them?

Is it okay if I label my images with their original size and then resize them, or should I first resize them and then label them? I mean do I need to recalibrate my labels if I resized my images?
1
vote
1answer
67 views

How to quickly change hand-drawn shapes to symmetrical polished shapes?

Given a hand-drawn shape, I'd like to generate the corresponding symmetrical polished shapes such as circle, rectangle, triangle, trapezoid, square, parallelogram, etc. A short video demonstration ...
1
vote
1answer
63 views

YOLOv3 Synthetic Data Training

Suppose we want to train a model to detect various objects. Let's say we have training data of those objects in various backgrounds along with their bounding boxes. Basically these objects have been ...
1
vote
0answers
137 views

How can I recognise the name of a molecule given an image of its structure?

I want to recognize the name of the chemical structure from the image of the chemical structure. For example, in the image below, it is a benzene structure, and I want to recognize that it is benzene ...
1
vote
0answers
17 views

learning object recognition of primitive shapes through transfer learning problem

Question on transfer learning object classification (MobileNet_v2 with 75% number of parameters) with my own synthetic data: I made my own dataset of three shapes: triangles, rectangles and spheres. ...
1
vote
1answer
369 views

Is Mask R-CNN suited to solve a multi-class classification problem where the classes are related?

I want to create a model to solve a multi-class classification problem. Here are more details about my problem. Every picture contains only one object The background is very simple All objects ...
1
vote
0answers
26 views

Where to find pre-trained models for multi-camera people tracking?

I need to build a multi-camera people tracking system and I have no idea how to start. I read ML for Dummies and I've watched a lot of youtube classes/conferences and read a lot of articles about ML/...
1
vote
0answers
66 views

Understanding average precision (AP) in measuring object detector performance

I am trying to understand the average precision (AP) metrics in evaluating the performance of deep-learning based object detection models. Suppose we have the following ground true (four objects ...
1
vote
0answers
14 views

Training Haar Cascade model with grey vs color images

Most examples if not all, are models that have been trained with images that are turned grey. Does this mean that models only detect edges? Why wouldnt you want to keep color so that model could learn ...
1
vote
0answers
134 views

Will balancing dataset of images for object detection for a single-shot OD (Yolov3-spp) by cropping lower the quality of the model?

I'm trying to work out an approach to balancing my dataset, which is a subset of a google openimages - some classes are represented orders of magnitude more than others and I am hesistant to simply ...
1
vote
1answer
899 views

Add training data to YOLO post-training

(Cross-posting here from the data science stack exchange, as my question didn't get any replies. I hope it's okay!) I've been playing around with YOLOv3 and obtaining some good results on the ~20 ...
1
vote
0answers
329 views

How does the target output of a Single Shot Detector (SSD) look like?

According to the paper SSD: Single Shot MultiBox Detector, for each cell in a feature map k boxes are acquired and for each box we get $c$ class scores and $4$ offsets relative to the original default ...
1
vote
0answers
83 views

Object recognition by two or more traits that are orthogonal (informally speaking)

I would really appreciate if someone could comment the following method of training neural nets providing them with some meta data (Making them more color prone only if needed, whereas now they're ...
1
vote
0answers
40 views

FIlling space with empty bounding box

I'm detecting objects on images. I want to detect up to 10 objects, however, I'm not sure how to deal with the situation, where only one object is present. Should I fill the remaining spaces in the ...
1
vote
0answers
239 views

Transfer learning from model trained in a similar dataset

I am currently working on a defect detection algorithm but I only have a few samples of defects.I googled for defect detection datasets and I found this one: http://resources.mpi-inf.mpg.de/...
0
votes
0answers
20 views

Consecutive frames can be discarded when training an SSD/YOLO?

Let's say I have a number of videos, and I want to train an SSD/YOLO (or FRCNN) to detect objects. In the case of a large amount of videos, there will be a lot of frames extracted and transferred to ...