SNIPER
Multi-scale detector
An efficient multi-scale object detection and inference algorithm using memory-efficient design and batch normalization
SNIPER / AutoFocus is an efficient multi-scale object detection training / inference algorithm
3k stars
82 watching
449 forks
Language: Python
last commit: over 3 years ago
Linked from 2 awesome lists
Related projects:
Repository | Description | Stars |
---|---|---|
| A deep learning library for image segmentation and object detection using PyTorch. | 1,054 |
| A software framework for training object detection models on multiple GPUs using the Caffe deep learning framework | 193 |
| A package providing advanced explainability methods for deep learning models in computer vision | 10,781 |
| A Python package for building and deploying computer vision models with PyTorch | 614 |
| A Python-based implementation of a single-shot object detection algorithm using Mask R-CNN architecture. | 339 |
| A deep learning method for optimizing convolutional neural networks by reducing computational cost while improving regularization and inference efficiency. | 18 |
| A toolkit for building and deploying deep learning models in computer vision | 5,850 |
| A deep learning framework for medical image segmentation using multi-scale guided attention mechanisms to improve accuracy and reduce irrelevant information. | 461 |
| Real-time object detection using deep learning and a large dataset of classes | 1,184 |
| Trains a bottom-up attention model using Faster R-CNN and Visual Genome annotations for image captioning and VQA tasks | 1,438 |
| Develops an object segmentation algorithm to detect camouflaged objects in images with varying backgrounds and contexts. | 20 |
| An image processing project implementing edge-based defocus blur estimation with adaptive scale selection | 18 |
| An implementation of Faster R-CNN object detection in PyTorch, modified from DenseCap. | 85 |
| A tool for defining and running machine learning experiments for image segmentation in Python. | 53 |
| A PyTorch implementation of the YOLO (You Only Look Once) v2 object detection algorithm with flexible configuration and parallel training capabilities. | 440 |