I'm in the process of reworking the ASAM database. Excerpted, it looks like this: 4155 PIRATES BULK CARRIER GULF OF ADEN: Bulk carrier fired upon 3 Aug 09 at 1500 UTC while underway in position 13-46.5N 050-42.3E. Ten heavily armed pirates in two boats fired upon the vessel underway. The pirates failed to board the vessel due to evasive action taken by the master. All crew and ship properties are safe (IMB). 4156 PIRATES CARGO SHIP NIGERIA: Vessel (SATURNAS) boarded, …

I'm trying to train a siamese network which contains a CNN and an embedding layer at the end to yield 2 similar (close) vectors for 2 images of the same person. I'm using the LFW_Cropped dataset, and some custom made generators. The generators are tested and returns batches of 50% 50% Same and Different pairs of images with the correct label. The labels for same and different outcome are: SAME = 1 -> (named as 'yes' in my code) DIFFERENT …

My goal is to use pretrained VGG16 to compute the feature vectors excluding the top layer. I want to compute embedding(no training involved) per image one by one rather than feeding batches to the network as this network is just to compute embedding and not classification. In batch training I understand the importance of batch normalisation but for a single image should I normalise it pixel wise? All I can think is maybe its useful to reduce the importance of …

Reading this paper on one-shot learning "Siamese Neural Networks for One-shot Image Recognition" I was introduced to the idea of Siamese Neural Networks. What I did not fully grasp was what they meant by this line: This objective is combined with standard backpropagation algorithm, where the gradient is additive across the twin networks due to the tied weights. Firstly, how exactly are they tied? Bellow, I believe I've provided the formula by which they compute the gradient. T is the …

I used to believe in k-way-n-shot few-shot learning, k and n (number of classes and samples from each class respectively) must be the same in train and test phases. But now I come across a git repository for few-shot learning that uses different numbers in the train and test phase : parser.add_argument('--dataset') parser.add_argument('--distance', default='l2') parser.add_argument('--n-train', default=1, type=int) parser.add_argument('--n-test', default=1, type=int) parser.add_argument('--k-train', default=60, type=int) parser.add_argument('--k-test', default=5, type=int) parser.add_argument('--q-train', default=5, type=int) parser.add_argument('--q-test', default=1, type=int) Are we allowed to do so?

I‘m looking for a minimal applied example for the implementation of a (one shot) Siamese Network, preferably in Keras. I‘m well aware of the various data science online pages and the respective examples and exercises that can be found there. However, so far I did not found an instructive source there. I would be thankful if someone could point me to some github source or if someone could share some code or other sources, which provide a sound example on …

I am training my SSD-300 model for which I have resized images to 300x300. I am using the default settings as mentioned in github repo: https://github.com/balancap/SSD-Tensorflow The loss is unstable while training. I tried training it till 50,000 training steps. The current mAP that I am getting is 0.26(VOC 2007) and 0.24 (VOC 2012) Train set: 1500 images Test: 300 images Current parameters: !python train_ssd_network.py --dataset_name=pascalvoc_2007 --dataset_split_name=train --model_name=ssd_300_vgg --save_summaries_secs=60 --save_interval_secs=600 --weight_decay=0.00004 --optimizer=adam --learning_rate=0.01 --batch_size=2 --gpu_memory_fraction=0.9 --learning_rate_decay_factor=0.94 -num_classes=3 --checkpoint_exclude_scopes =ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box --eval_training_data=True …

I have a dataset with a lot of classes (~10000+) but few examples by classes (~15-). I want to classify these classes, but there are some specificities. My examples provide from a video stream. Currently, I use a detector (like YOLO) to do object detection and then I apply a siamese/triplet network to obtain embeddings. Finally, I classify by using a Kneighbor classifier with one neighbor. The real specificity, is that I want to detect when a picture provided by …

Being new to deep learning I am somewhat struggling to grasp the idea of one shot learning. Let us say I have a class to detect which didn't exist in training dataset such as COCO or Image NET. Can I train model for that class using only image or the training set must be large as for YOLO or RCNNs?

I'm currently looking into one-shot learning and I wonder if there are any good sources/tutorials out there, which demonstrate one-shot learning in a (more or less) "hands-on" way? Also some hints to indroductory literature to one-shot learning would be highly welcome! Are there any books which cover the topic? I'm not interested in standard ML homepages which pop up on the first page of a Google search. I'm looking for recommendations of people who are somewhat familiar with the topic. …

I have a preprocessing pipeline that includes replacing missing values and onehotencoding for the categorical variables. When I try to use my model on the test set, it explains that the number of columns it expects differs. This is due to one hot encoding One option I considered was passing the full dataset into the pipeline and then seperating into test and split. However, this causes data leakage as the missing values it capturing values from the testset. Please let …