Jul 22, 2019 · featured post

New research: Deep Learning for Image Analysis

We discussed this research as part of our virtual event on Wednesday, July 24th; you can watch the replay here!

Convolutional Neural Networks (CNNs or ConvNets) excel at learning meaningful representations of features and concepts within images. These capabilities make CNNs extremely valuable for solving problems in the image analysis domain. We can automatically identify defects in manufactured items, reducing costs associated with quality assurance processes; we can now infer depth information and reconstruct 3D maps from 2D images without additional metadata, giving new potential to urban planning as well as entertainment experiences; we can perform pixel level separation of objects in images and video with applications ranging from public safety to medical robotics; and we can perform “super resolution” on photo images (upscale an image up to 10x) with a trained deep-learning system reconstructing, filling in, and thus sharpening images with information that would be omitted and lost using a standard digital zoom. It’s a long list.

Historically, CNN models have been manually designed (e.g Vgg16, ResNet50, InceptionV4 etc). While these models have demonstrated accuracy beyond human capabilities for many tasks, they tend to be large (parameters, floating point operations) and inefficient. More recently, approaches which enable the automatic discovery of model architectures (a field that is becoming known as neural architecture search or autoML) has shown promise for the discovery of high accuracy, high efficiency models. These autoML architectures, when applied to a broad range of tasks (e.g., object detection, segmentation, image generation et.c), are a recommended approach to unlock even more value for businesses.

The landscape of competitive deep learning models for image analysis. We observe that neural architecture search (AutoML) has resulted in the discovery of highly accurate model architectures at a fraction of the costs (trainable parameters). These models are able to achieve more with less.

In our upcoming report, we cover a variety of practical considerations relevant to deploying deep learning models for image analysis. We begin by discussing a set of image analysis tasks and offer recommendations on which models to use and why. We also explore a set of approaches for interpreting models, tools/libraries, and ethical concerns that arise.


Along with this report, we have created a prototype - ConvNet Playground - an interactive visualization for exploring Convolutional Neural Networks, applied to the task of semantic image search. The prototype allows users to select search configurations (datasets, models, layers, distance metrics) and interactively explore how these impact search query performance. It allows users to ask questions such as: How does semantic search (with pretrained models) perform for given datasets? How well does each model/layer configuration capture “semantic meaning” for a given dataset? For a given search query, how does search performance compare for each model/layer configuration? What type of features/patterns are detected by various layers in a pretrained model?

The approach in this prototype is implemented in two stages (i.) we extract features from all images in our datasets using a pre-trained CNN (think VGG16, InceptionV3, etc. pre-trained on imageNet) (ii.) We compute similarity as a measure of the distance between these features.

To implement semantic search, we start by extracting features from images in our dataset using pre-trained CNN models. Similarity is computed as the distance between these features.

Interested in a walkthrough of this prototype and learning more about other recent advances in deep learning for image analysis? We discussed this research as part of our virtual event on Wednesday, July 24th; you can watch the replay here!

Read more

Aug 28, 2019 · newsletter
Jul 17, 2019 · post

Latest posts

May 5, 2022 · post

Neutralizing Subjectivity Bias with HuggingFace Transformers

by Andrew Reed · Subjective language is all around us – product advertisements, social marketing campaigns, personal opinion blogs, political propaganda, and news media, just to name a few examples. From a young age, we are taught the power of rhetoric as a means to influence others with our ideas and enact change in the world. As a result, this has become society’s default tone for broadcasting ideas. And while the ultimate morality of our rhetoric depends on the underlying intent (benevolent vs. more
Mar 22, 2022 · post

An Introduction to Text Style Transfer

by Andrew Reed · Today’s world of natural language processing (NLP) is driven by powerful transformer-based models that can automatically caption images, answer open-ended questions, engage in free dialog, and summarize long-form bodies of text – of course, with varying degrees of success. Success here is typically measured by the accuracy (Did the model produce a correct response?) and fluency (Is the output coherent in the native language?) of the generated text. While these two measures of success are of top priority, they neglect a fundamental aspect of language – style. more
Jan 31, 2022 · post

Why and How Convolutions Work for Video Classification

by Daniel Valdez-Balderas · Video classification is perhaps the simplest and most fundamental of the tasks in the field of video understanding. In this blog post, we’ll take a deep dive into why and how convolutions work for video classification. Our goal is to help the reader develop an intuition about the relationship between space (the image part of video) and time (the sequence part of video), and pave the way to a deep understanding of video classification algorithms. more
Dec 14, 2021 · post

An Introduction to Video Understanding: Capabilities and Applications

by Daniel Valdez Balderas · Video footage constitutes a significant portion of all data in the world. The 30 thousand hours of video uploaded to Youtube every hour is a part of that data; another portion is produced by 770 million surveillance cameras globally. In addition to being plentiful, video data has tremendous capacity to store useful information. Its vastness, richness, and applicability make the understanding of video a key activity within the field of computer vision. more
Sep 22, 2021 · post

Automatic Summarization from TextRank to Transformers

by Melanie Beck · Automatic summarization is a task in which a machine distills a large amount of data into a subset (the summary) that retains the most relevant and important information from the whole. While traditionally applied to text, automatic summarization can include other formats such as images or audio. In this article we’ll cover the main approaches to automatic text summarization, talk about what makes for a good summary, and introduce Summarize. – a summarization prototype we built that showcases several automatic summarization techniques. more
Sep 21, 2021 · post

Extractive Summarization with Sentence-BERT

by Victor Dibia · In extractive summarization, the task is to identify a subset of text (e.g., sentences) from a document that can then be assembled into a summary. Overall, we can treat extractive summarization as a recommendation problem. That is, given a query, recommend a set of sentences that are relevant. The query here is the document, relevance is a measure of whether a given sentence belongs in the document summary. How we go about obtaining this measure of relevance varies (a common dilemma for any recommendation system). more

Popular posts

Oct 30, 2019 · newsletter
Exciting Applications of Graph Neural Networks
Nov 14, 2018 · post
Federated learning: distributed machine learning with data locality and privacy
Apr 10, 2018 · post
PyTorch for Recommenders 101
Oct 4, 2017 · post
First Look: Using Three.js for 2D Data Visualization
Aug 22, 2016 · whitepaper
Under the Hood of the Variational Autoencoder (in Prose and Code)
Feb 24, 2016 · post
"Hello world" in Keras (or, Scikit-learn versus Keras)


In-depth guides to specific machine learning capabilities


Machine learning prototypes and interactive notebooks


A usable library for question answering on large datasets.

Explain BERT for Question Answering Models

Tensorflow 2.0 notebook to explain and visualize a HuggingFace BERT for Question Answering model.

NLP for Question Answering

Ongoing posts and code documenting the process of building a question answering model.

Interpretability Revisited: SHAP and LIME

Explore how to use LIME and SHAP for interpretability.

Cloudera Fast Forward Labs

Making the recently possible useful.

Cloudera Fast Forward Labs is an applied machine learning research group. Our mission is to empower enterprise data science practitioners to apply emergent academic research to production machine learning use cases in practical and socially responsible ways, while also driving innovation through the Cloudera ecosystem. Our team brings thoughtful, creative, and diverse perspectives to deeply researched work. In this way, we strive to help organizations make the most of their ML investment as well as educate and inspire the broader machine learning and data science community.

Cloudera   Blog   Twitter