Stack Overflow
Stack Overflow

Questions tagged [tensorflow-ssd]

Tensorflow Single Shot Detector(SSD) is a way to identify multiple objects within an image. Use with the [tensorflow] tag

Originally described in this paper

We present a method for detecting objects in images using a singledeep neural network. Our approach, named SSD, discretizes the output space ofbounding boxes into a set of default boxes over different aspect ratios and scalesper feature map location.

22 questions
6
votes
1 answer

SSD mobilenet model does not detect objects at longer distances

I have trained an SSD Mobilenet model with custom dataset(Battery). Sample image of the battery is given below and also attached the config file which I used to train the model. When the object is closer to the camera(tested with webcam) it…
Dinesh
  • 1,135
  • 2
  • 15
  • 22
4
votes
0 answers

Error: Tensor must have a shape comprised of positive integers but got shape [100,]

To get help from the community, we encourage using Stack Overflow and the tensorflow.js tag. Browser version not in the browser, using Node command Describe the problem or feature request I am using tensorflow.js of coco-SSD for mobilenet…
Mithilesh
  • 223
  • 2
  • 13
4
votes
3 answers

Converting SSD to frozen graph in tensorflow. Which output node names must be used?

I trained SSD using TensorFlow Object Detection API as described here. It produces a ckpt, meta and index file. In order to run it on my images I tried to check the demo code. It requires that the model be converted to frozen graph. I tried to…
Srikrishan Bhat
  • 45
  • 1
  • 5
2
votes
0 answers

Is it possible to use the gpu of mobile phone

I have tried android demo of tensorflow on my phone, And successfully used my own ssd_mobilenets model trained by object_detection API. It said ssd_mobilenets can run on android phone in real time, But it runs a little slow on my mobile phone. It…
Seven.L
  • 93
  • 2
  • 8
1
vote
1 answer

SSD anchor box scale relative to input image size or feature map

I have done a lot of reading on SSD paper, but i couldn't figure out when they mention a scale of Smin 0.2 to Smax 0.9, do they mean in relation to the input image size or in relation to the feature map size at that layer. Can someone help clarify…
Pratik Khadloya
  • 12,509
  • 11
  • 81
  • 106
1
vote
0 answers

OpenCV dnn exception SSD Mobilenetv2

Using transfer learning, I trained SSD MobileNetV2 (ssd_mobilenet_v2_coco.config) model in TensorFlow (tensorflow-gpu==1.15.0). After freezing the graph (.pb) using TensorFlow API Python script (export_inference_graph.py), I created a text graph…
turing
  • 11
  • 1
1
vote
0 answers

create_tf_record without trimaps

I am following the "Towards Data Science" guide https://towardsdatascience.com/custom-object-detection-using-tensorflow-from-scratch-e61da2e10087 I ran into an issue on step 6 where I need to create a tf record. The create_tf_record.py script…
Ivan Viti
  • 271
  • 2
  • 11
1
vote
0 answers

Error while interpreting .tflite file of ssd_resnet50 object detection model

The error is Kernal Restarting: The kernel appears to have died. It will restart automatically. Am following the steps as mentioned here. The kernel crashes on executing interpreter.invoke() - in the code below. Am running the code through Jupyter…
Malgo
  • 1,871
  • 1
  • 17
  • 30
1
vote
2 answers

Training SSD-MOBILENET V1 and the loss does not deacrease

I'm new in everithing about CNN and tensorflow. Im training a pretrained ssd-mobilenev1-pets.config to detect columns of buildings, about one day but the loss is between 2-1 and doesnt decrease since 10 hours ago. I realized that my input images…
Mishel Mejia
  • 11
  • 1
  • 2
1
vote
1 answer

Training in SSD Implementation in Keras halts after few iteration without any output or error

After few iterations of first epoch the training process halts without any output or error message. SSD implementation in Keras was used from https://github.com/rykov8/ssd_keras base_lr = 3e-4 #optim = keras.optimizers.Adam(lr=base_lr) optim =…
Deepesh Lekhak
  • 13
  • 4
1
vote
1 answer

Tensorflow Object Detection API on Android

I'd like to swap out the multibox_model.pb being used by TensorFlowMultiBoxDetector.java in Google's Tensorflow Detect Sample App with the mobilenet frozen_inference_graph.pb included in the object detection API's model zoo. I've ran the…
matteo411
  • 390
  • 2
  • 6
0
votes
1 answer

How to define prob_threshold to avoid double counting during object detection?

I am developing an object detection application using SSD model and I have defined the bounding box and the prob_threshold, when I run the code I realise that the model double count person in frame. Please see below my code ## Setting Pro_threshold…
Godspower
  • 87
  • 2
  • 6
0
votes
1 answer

How to chose the number of anchors

How I can implement a single shot detector with only one anchor box per feature map cell? I have used ssd mobilenet coco 2018 v1 and I have changed the anchor generator part in config file I.e num layers = 1 in ssd anchor generator. Will that be…
user9631198
  • 1
  • 1
0
votes
2 answers

Tensorflow Object Detection API: Training gets stuck at step=0 for ssd + mobilenetv2 with custom data

I wanted to do transfer learning using a ssd + mobilenetv2 model with my own images. I have only one class. The images were downloaded from OpenImageDataSet. I used tensorflow's object detection API. But the training stuck at step = 0. I verified…
jackyvr
  • 11
  • 1
  • 4
0
votes
0 answers

tensorflow object detection api (ssd + mobilenet) for ocr (detection and reading). Bad for long symbol sequences

I am trying to learn Tensorflow Object Detection API (SSD + MobileNet architecture) on the example of reading sequences of Arabic numbers. Generated images with random sequences of numbers of different lengths - from one digit to 20 were fed to the…
lenkele
  • 1
1
2