Tracker

TCTrack: Temporal Contexts for Aerial Tracking (CVPR2022)

Ziang Cao and Ziyuan Huang and Liang Pan and Shiwei Zhang and Ziwei Liu and Changhong Fu In CVPR, 2022. [paper] Abstract Temporal contexts among consecutive frames are farfrom being fully utilized in existing visual trackers. In this work, we present TCTrack, a comprehensive framework to fully exploit temporal contexts for aerial tracking. The temporal contexts are incorporated at two levels: the extraction of features and the refinement of similarity maps. Specifically, for feature extraction, an online temporally adaptive convolution […]

Read more
Loading

Populate library namespace without incurring immediate import costs

lazy-loader makes it easy to load subpackages and functions on demand. Motivation Allow subpackages to be made visible to users without incurring import costs. Allow external libraries to be imported only when used, improving import times. For a more detailed discussion, see the SPEC. Installation pip install -U lazy-loader Usage Lazily load subpackages Consider the __init__.py from scikit-image: subpackages = [ …, ‘filters’, … ] import lazy_loader as lazy __getattr__, __dir__, _

Read more
Twitter

Twitter Bot to track updates in Metaculus questions

This bot monitors some Metaculus questions and posts a tweet when the predictions have changed significantly. The bot is currently deployed on Google Cloud Functions (deploy.sh) and runs every 6 hours (schedule.sh). The main function that runs the bot is in main.py. Contributions Contributions are very welcome, please open an issue or create a Pull Request. GitHub View Github    

Read more
Image Processing

Using pixel classification and some Image Processing Techniques to detect bin in images

This project focuses on using the logistic regresion based image pixel classification model and image reinforcement methods to segment and augment images and then detected the positions of the bins in different images. As bins may appear in different places, it is necessary to design a classification algorithm that can segment the bin from a complex scenario. My method achieved a reliable outcome on detecting the location of bins in different scenarios. The techinical approach and methmetical formulation are all […]

Read more
Deep Learning

Boosting Monocular Depth

(NEW!) Boost Your Own depth with our new repo We present a stand-alone implementation of our Merging Operator. This new repo allows using any pair of monocular depth estimations in our double estimation. This includes using separate networks for base and high-res estimations, using networks not supported by this repo (such as Midas-v3), or using manually edited depth maps for artistic use. This will also be useful for scientists developing CNN-based MDE as a way to quickly apply double estimation […]

Read more
Cloud

Google Cloud Platform (GCP) configuration scanning tool

Cloudimized is a Google Cloud Platform (GCP) configuration scanning tool. It allowsmonitoring changes of selected resources. Cloudimized performs similar function as Oxidized, but for Cloud environment. Overview Cloudimized periodical scans of GCP resources via API calls and dumps them into yaml files. Files are tracked in Git,so every configuration change is being tracked.It gathers additional information for each change: changer’s identity, related Terraform runs(optionally), identify change ticket number (optionally).Optionally it sends Slack notifications for each change. Features Project discovery across […]

Read more
OCR

Comparison-of-OCR (KerasOCR, PyTesseract,EasyOCR)

OCR (Optical Character Recognition) is a technology that enables the conversion of document types such as scanned paper documents, PDF files or pictures taken with a digital camera into editable and searchable data. OCR creates words from letters and sentences from words by selecting and separating letters from images. Requirements pip install -r requirements.txt Usage Also you can check the result by one by like: You have to initilaze your object. ocr=OCR(image_folder=”test/”) After that, for keras ocr: for easyocr: ocr.easyocr_model_works() […]

Read more
Optimization

Optimizing Protein Structure Prediction Model Training and Inference on GPU Clusters

Optimizing Protein Structure Prediction Model Training and Inference on GPU Clusters FastFold provides a high-performance implementation of Evoformer with the following characteristics. Excellent kernel performance on GPU platform Supporting Dynamic Axial Parallelism(DAP) Break the memory limit of single GPU and reduce the overall training time DAP can significantly speed up inference and make ultra-long sequence inference possible Ease of use Huge performance gains with a few lines changes You don’t need to care about how the parallel part is implemented […]

Read more
FastAPI

Simple example of FastAPI + Celery + Triton for benchmarking

You can see the previous work from: with Triton Server without Triton Server Benchmark Results See Benchmark Results Preparation 1. Setup packages Install Anaconda and execute the following commands: $ make env # create a conda environment (need only once) $ source init.sh # activate the env $ make setup # setup packages (need only once) 2. Train a CNN model (Recommended on GPU) 3. Check the model repository    

Read more
Transformer

Official Python package for Deep Kernel Shaping (DKS) and Tailored Activation Transformations (TAT)

This Python package implements the activation function transformations andweight initializations used Deep Kernel Shaping (DKS) and Tailored ActivationTransformations (TAT). DKS and TAT, which were introduced in the DKS paper andTAT paper, are methods constructing/transforming neural networks to make themmuch easier to train. For example, these methods can be used in conjunction withK-FAC to train deep vanilla deep convnets (without skip connections ornormalization layers) as fast as standard ResNets of the same depth. The package supports the JAX, PyTorch, and TensorFlow […]

Read more
1 … 308 309 310 311 312 … 1,023