Questions tagged [object-detection]

1

votes
0

answer
172

Views

Google TensorFlow Object Detection API on Windows - Labelling Images

I have been using Labellmg to create the xml files in PASCAL VOC format. There is a prebuilt binary version which makes it really easy to start drawing bounding boxes around objects in the images. https://github.com/tzutalin/labelImg At times, the bounding boxes end up covering unwanted pixels. I...
kannan k
1

votes
0

answer
171

Views

How CNNs achieve spatial invariance despite having a Fully Connected layer at the end for classification?

An image of a dog should be learnt by the network as being a dog irrespective of the spatial position and or rotation (maybe even deformation?). My question is how CNNs are able to achieve this? The convolution operations on input image successfully detect a feature that is characteristic of a parti...
James Howlett
1

votes
1

answer
415

Views

Tensorflow Object detection development in IOS app - nonmaxsuppressionV2

I am trying to detect object using ssd_mobilenet_v1_coco model. My own trained model file .pb file is used for detection. After successful build , click run button and I got the below error. 'Not found: Op type not registered 'NonMaxSuppressionV2' in binary running on IPhone. Make sure the Op and Ke...
Balaganesh R
1

votes
0

answer
40

Views

Tensorflow - running python files in Spyder

to give a bit of background I'm trying to use TensorFlow object detection api. I have been following this tutorial https://www.youtube.com/watch?v=MoMjIwGSFVQ 15 minutes into it he uses Spyder, and runs the python files on it. I've never used Spyder before, and for some reason it's not running my f...
BBKING
1

votes
0

answer
569

Views

Tensorflow Object Detection using Tensorflow Mobile

I am trying to use a custom model in the TF-Detect Android Demo. model: ssd_mobilenet_v1_coco The model is trained on 8 classes. After exporting the model I've optimised it using Tensorflow-Mobile bazel-bin/tensorflow/tools/graph_transforms/transform_graph \ --in_graph=frozen_inference_graph.pb \ --...
Krishnendu S. Kar
1

votes
1

answer
48

Views

Is there a way to only detect objects with width > 200px in tensorflow?

Is there a way filter object detection with width? Assuming that I have to detect apples with WIDTH > 100 Pixels. Is there a build-in tensorflow function to do that? or do I have to condition the code myself before drawing the bounding box?
Muhammad Hamza Javed
1

votes
1

answer
153

Views

Trained model detects almost everything as one class after a long training

I trained a custom person detector using Tensorflow and Inception's pretrained model then after a few thousands of step and an average of 2-1 loss, I've stopped the training and tested it with a live video. The result was quite good and only gets few false positives. It can detect some person but no...
1

votes
1

answer
684

Views

Error running tensor flow model `TypeError: __init__() got an unexpected keyword argument 'file'`

I'm trying to install tensorflow Object Detection API. I have followed all the installations steps as of here. However, when I tried running python3 object_detection/builders/model_builder_test.py, I got this error: Traceback (most recent call last): File 'object_detection/builders/model_builder_tes...
Stanley Nguyen
1

votes
1

answer
228

Views

Tensorflow empty model.ckpt

I try to design my own model detector on tensorflow with objectdetector API. I run the running pet example, but it tells me this: To speed up the training process, it is recommended that users re-use the feature extractor parameters from a pre-existing object classification or detection checkpoint....
dev dev
1

votes
1

answer
654

Views

Error running export_inference_graph.py in Object Detection API

When i run export_inference_graph.py with the correct paths from the object_detection folder, or from the models/research folder like this: python export_inference_graph \ # object_detection/ (if from /research) --input_type image_tensor \ --pipeline_config_path path/to/ssd_inception_v2.config \...
GustavZ
1

votes
0

answer
411

Views

How to Optimize a Trained Frozen Model for Inference?

My goal is to decrease the size and complexity of a pre-trained Model (a Tensorflow Frozen Graph as Protobuf .pb file) as far as possible to make Inference (in my case realtime object detection using Webcams) as fast as possible. (See my project repo for more information: https://github.com/GustavZ/...
GustavZ
1

votes
0

answer
649

Views

Object detection boxes are lost at second evaluation step

I'm a beginner with Tensorflow 1.4.0 and I'm trying to perform my first training + evaluation process on an object detection model. What I'm seeing is something weird when looking at the output of the evaluation steps. Here is the steps I made. First, it's worth to say that my goal is to detect two...
TheUnexpected
1

votes
1

answer
1k

Views

tensorflow object detection API(Calculate Car speeds.)

I used tensoflow object detection API to count the number of cars detected. But now i want to calculate the speed of all the cars detected. My question is is there any way to do this using tensorflow object detection API.
user3432888
1

votes
0

answer
259

Views

Obj and No Obj fields in YOLOv2 darknet always 0

I am training YOLOv2 model on custom data. It has 6 classes of data. I follow AlexyAB's darknet to train for custom training. The dataset I am using is COQE dataset mentioned in this paper. The problem I am facing is whenever I start training the Obj and No-Obj attribute goes to zero. In the cfg fil...
Perseus14
1

votes
0

answer
65

Views

How to conduct the object detection on sequence data/text (1D matrix)?

I tried to search for some Keras codes for the object detection. But almost all of them deal with images. But what I want to do is to conduct object detection on sequence data/text. Here's the detail: I have some sequences like this: (target region)(non-target region)(target region)(non-target regio...
Ying Shan
1

votes
1

answer
460

Views

tensorflow object_detection train.py error

I'm currently trying to retrain Google's Tensorflow Object Detection I followed the steps in sentdex's tutorial (and probably like many of you did). After I generated the tf_records and moved them in the directory models/research/object_detection, everything should be set for the training. To initi...
bogdan.georgescu.gbm
1

votes
1

answer
846

Views

Color Detection in specific object Tensorflow

I am implementing the Tensorflow object detection in on of my android app, I have followed the demo and tutorial of 'Tensorflow-for-poets' and successfully created a model with that. I need one help with this, I have this requirement of detecting the traffic signal, I have a dataset and I have creat...
Hardik Chauhan
1

votes
1

answer
1.9k

Views

faster-rcnn config file in tensorflow

I am using Google API for object detection in tensorflow to train and infer on a custom dataset. I would like to adjust the parameters of the config file to better suit my samples (e.g. no. of region proposals, size of ROI bbox, etc.). To do so, I need to know what each parameter does. Unfortunately...
Hafplo
1

votes
1

answer
1.3k

Views

Understanding and tracking of metrics in object detection

I have some questions about metrics if I do some training or evaluation on my own dataset. I am still new to this topic and just experimented with tensorflow and googles object detection api and tensorboard... So I did all this stuff to get things up and running with the object detection api and tra...
user8574993
1

votes
0

answer
70

Views

Change and use pre-trained model architecture

I am using Tensoflow object detection API for detection of objects.Currently, I am using MobileNet pre-trained model. Now I need to evaluate pre-trained model results under different forms of the model.To do that I need to do some slit changes to the model architecture. Is it possible?. If then How...
Chamod Pathirana
1

votes
0

answer
231

Views

How to understand RPN phrase in Faster-RCNN

Sorry, it is all by one question but relate to many small questions. I can't split them into seperated questions. For example, input picture size 960x640 Through VGG16 layer 13 Conv5_3, get feature_map 60x40x512 Do 3x3 convolution.     3.1 How 3x3 convolution compress the output above to 1x512...
Mithril
1

votes
1

answer
262

Views

TensorFlow Custom Object Detection Disappointing Result - Why?

I have just started TF Object Detection API two weeks ago, and manage to train a model to recognize a custom object, in my case, a Mecanum wheel. Here's the details: - No. of training images = 125 - All training images are around 500 x 500 (plus minus) - Transfer Learning - Model used = ssd_mobilen...
iSeeDeadPeople
1

votes
1

answer
388

Views

ERROR: Couldn't match files for checkpoint gs://obj-detection/train/model.ckpt

I run my detection model on google cloud ml and got this error while running the evaluation script. I found this link that mentioned about this issue, but it seems like the issue's till not be solved. Anyone knows how to fix this? Any helps would be appreciated. Thanks. ERROR 2018-02-04 12:53:10 -06...
Tung Le
1

votes
0

answer
250

Views

object detection api for gray scale images

I'm training a tf model to recognize various objects. I have many color images i use to train the model. however, in real life my app will encounter many images taken at gray-scale due to lighting conditions or camera capabilities.is there an easy way to tell TF to train the model both on the color...
asaf oron
1

votes
0

answer
230

Views

[Tensorflow]Different Graph architecture in COCO pre-trained model and re-train model

I have followed the doc: https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_locally.md to retrained the model of ssd_mobilenet_v1_coco_2017_11_17.tar.gz. I only modify the pipe config with - finetune to TRUE - set the model path - set the trained data path wit...
Dogee
1

votes
0

answer
41

Views

Is it possible to train a network based on the output of other neural network in TensorFlow?

I am trying to create a neural network that can process the input images for another neural network, say object detection. If possible, I try to take the advantage of TensorFlow Models (https://github.com/tensorflow/models). There are two options that I would like to try: Freeze object detection net...
rogerc
1

votes
0

answer
352

Views

Change rcnn_resnet_101.config file in Tensorflow object-detection API for high resolution training

I am referring to: (https://github.com/tensorflow/models/tree/master/research/object_detection). My training set consists of labelled images with boxes exported in xml format for training. The dimentions of my input training images are min_dimension: 2976 max_dimension: 4464. By default the config f...
Ajinkya
1

votes
2

answer
604

Views

TensorFlow Object Detection API training error

Hi there, I execute the following in my Anaconda prompt:(tensorflow) C:\Users\nicho\Documents\01_Machine_Learning\00_Lynda.com\Ex_Files_TensorFlow\models\research\object_detection> python train.py --logtostderr --train_dir=training/ --pipeline_config_path=training/ssd_mobilenet_v1_pets.config I am t...
Nicholas Zembashi
1

votes
1

answer
780

Views

How to not resize input image while running Tensorflow SSD's inference

From what I can understand from Single Shot Multibox Detector paper, it is a fully convolutional network. As such, it won't require the rescaling which tensorflow is doing (to 300x300) during inference. How can I remove this resizing during inference in tensorflow?
Krupa
1

votes
0

answer
349

Views

Tensorflow object detection API - training a model on multiple GPUs

My question refers to the relation between number_of_steps and batch_size when training a model using the Tensorflow object detection API. When I increase batch size. e.g. from 1 to 2, can I then half the number_of_steps and receive the same accuracy for the model?
Markus
1

votes
1

answer
204

Views

what does levelWeights mean in opencv detectMultiScale3?

Here is my implementation of HaarCascade where I trained my own classifier. import cv2 import numpy as np body_classifier = cv2.CascadeClassifier('C:\\Users\\Nemi\\MasteringComputerVision_V1.00\\Haarcascades\\trainedHuman.xml') image = cv2.imread('twn2.jpg') #####HEREEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE...
Nemi Bhattarai
1

votes
0

answer
415

Views

Tensorflow object detection: Change input resolution of model

I'm using Tensorflow's Object Detection API using the code provided in the repository to analyze high-resolution images (4K). I've tested a few pre-trained models provided in the Model Zoo. The models are configured to downscale the input image to a certain size specified in their config file. Due t...
Ignace Maes
1

votes
1

answer
651

Views

OpenCv SIFT brute force matching with multiple images

import cv2 import numpy as np MIN_MATCH_COUNT = 30 #Initiate Sift detector for brute force matching detector = cv2.xfeatures2d.SIFT_create() #FLANN Parameters(finds nearest neighbors) FLANN_INDEX_KDITREE=0 flannParam = dict(algorithm = FLANN_INDEX_KDITREE, tree=5) flann = cv2.FlannBasedMatcher(flann...
Chad French
1

votes
0

answer
115

Views

reshape with * Operator

can u please help me understand reshaping a matrix with the *operator? Tried googling and check the documentation of reshaping but couldnt find anything using it with *: matrix = matrix.reshape(*matrix.shape, 1) Following up that question can one of you help me understand the use of the line in this...
Felix Schäfer
1

votes
0

answer
23

Views

When I run train_net.py of fast rcnn,the code cracked

Here is when the code crack: I0303 14:17:35.769629 17352 net.cpp:774] Copying source layer loss_cls I0303 14:17:35.770619 17352 net.cpp:774] Copying source layer loss_bbox Solving... And then I use VS2013 to see where exist the error template const Dtype* Blob::gpu_data() const { CHECK(data_);//her...
hashiting
1

votes
0

answer
41

Views

TensorFlow trained model detecting every object as trained object

I have trained a faster_rcnn_resnet101_coco model for two objects Lion and Ostrich with 180-200 images for each object. After training, the model is detecting every object as either lion or ostrich irrespective of the object in the image. Example: It is detecting elephant as both lion and ostrich. C...
zakib
1

votes
1

answer
1.4k

Views

What size of image is better to have for the TensorFlow Object Detection API?

I want to know when training my own image dataset by using a pre-trained model (e.g: ssd_mobilenet_v1_coco, ssd_inception_v2_coco, faster_rcnn_inception_v2_coco, ...) what image size is better to resize the images before starting creating TF record and training? I am asking this because when I coll...
Bahramdun Adil
1

votes
0

answer
202

Views

For Faster-RCNN, is it possible to fine tune existing model if I want to also specify new anchor box scale and aspect ratio?

For Faster-RCNN, is it possible to fine tune existing model (say trained with Coco dataset) to detect a single class of objects, say traffic lights, if I want to also specify new anchor box scale and aspect ratio most suitable for detecting traffic signs? The reason is the default anchor boxes (for...
Jack Kwok
1

votes
0

answer
25

Views

Change the default values of aspect_ratios SSD to my own values

In the SSD512 object detection model the default values of aspect_ratios are : aspect_ratios_per_layer=[[1.0, 2.0, 0.5], [1.0, 2.0, 0.5, 3.0, 1.0/3.0], [1.0, 2.0, 0.5, 3.0, 1.0/3.0], [1.0, 2.0, 0.5, 3.0, 1.0/3.0], [1.0, 2.0, 0.5, 3.0, 1.0/3.0], [1.0, 2.0, 0.5], [1.0, 2.0, 0.5]], But i want to have...
deeeepnet
1

votes
0

answer
56

Views

How does the number of object detection classes affect performance in Tensorflow?

I'm using TensorFlow's object detection API. I trained my custom model from SSD_MOBILENET_COCO_2 that originally had 90 detection classes. My custom model only has 1 detection class and performance seems to be similar. Is there any relationship between the number of classes and the performance of th...
David Papp

View additional questions