A command-line utility that uses convolutional neural networks to search and filter videos based on objects and places that appear in them.
Thingscoop is a command-line utility that uses convolutional neural networks to semantically analyze videos. It allows users to search, filter, and describe video content based on objects and scenes that appear in them, automating the process of video content indexing and retrieval. The tool solves the problem of manually reviewing long videos to find specific moments by leveraging pre-trained deep learning models.
Developers, researchers, and media professionals who need to programmatically analyze or manipulate video content based on visual elements, such as those working in video editing, content moderation, or multimedia research.
Thingscoop provides a straightforward, scriptable interface for video content analysis using advanced deep learning models, eliminating the need for manual tagging or custom model training. Its unique selling point is the combination of a flexible query language with frame-accurate semantic search capabilities.
Search and filter videos based on objects that appear in them using convolutional neural networks
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Supports a basic query language with logical operators (AND, OR, NOT), enabling precise searches like 'sky && !ocean' for targeted video filtering without manual tagging.
Offers both object recognition (ImageNet) and scene recognition (MIT Places) models, allowing users to choose based on whether they need to identify objects or settings in videos.
The preview command plays videos while displaying recognized labels on each frame, providing immediate visual feedback for model accuracy and analysis results.
Uses models from 2015 (VGG and GoogLeNet), which are less accurate and efficient than modern architectures like YOLO or Transformers, and the project hasn't been updated to include newer models.
Requires manual setup of Caffe with Python bindings, plus ffmpeg and imagemagick, involving multiple steps and environment variable configuration that can fail on modern systems.
Last updated in 2015 with no recent activity, meaning it lacks bug fixes, compatibility updates, and support for current operating systems or libraries.