$$
{{ $t($store.state.user.experience_value_in_dollars) }}
Junior
{{ $t($store.state.user.experience_search_name) }}
0
jobs
ML Computer Vision Engineer
Alexander Spanellis
,
OXFORD, United Kingdom
Experience
Other titles
Skills
I'm offering
My skills lie within multiple disciplines within Deep Learning space, from real-time high speed detection/segmentation and recognition implementations to time-series classification and prediction. In addition, I have worked in application domains in structured light depth estimation, as well as human trajectory prediction and classification.
I have experience in both creating and designing custom Deep Learning Architectures, as well as in leveraging state of the art API’s to rapidly develop models to suit the requirements of the model and associated product. The majority of my experience involves working with limited datasets. As a result, I have utilised automated data augmentation and generation techniques to build high performance models within these constraints.
My interests lie in state of the art applications of Machine/Deep Learning techniques and their integration into real world solutions. I find particular satisfaction in implementing cutting edge research within the domain I am working, with the aim of pushing these concepts into real world applications.
I have experience in both creating and designing custom Deep Learning Architectures, as well as in leveraging state of the art API’s to rapidly develop models to suit the requirements of the model and associated product. The majority of my experience involves working with limited datasets. As a result, I have utilised automated data augmentation and generation techniques to build high performance models within these constraints.
My interests lie in state of the art applications of Machine/Deep Learning techniques and their integration into real world solutions. I find particular satisfaction in implementing cutting edge research within the domain I am working, with the aim of pushing these concepts into real world applications.
Links for more
Once you have created a company account and a job, you can access the profiles links.
Language
English
Fluently
Ready for
Larger project
Ongoing relation / part-time
Available
My experience
2019 - ?
project
MULTIVARIATE TIME SERIES CLASSIFICATION
Bitcoin time series classification using neural architecture search to build CNN-LSTM attention based recurrent architecture to estimate the position to take A position based on the fluctuations in the cryptos value.
Data Science, Deep learning, Software development
2018 - ?
job
MACHINE VISION ENGINEER
WheelRight.
∠ Focus on machine vision domain.
∠ Project based work with a focus on real world ML implementations.
∠ Developing robust models which are product ready while working within resource constraints.
∠ Worked on projects across machine vision domain from structured light regression and detection
to image reconstruction/generation techniques.
∠ Responsible for and supervised the complete end to end development the models and system
implementation.
∠ Project based work with a focus on real world ML implementations.
∠ Developing robust models which are product ready while working within resource constraints.
∠ Worked on projects across machine vision domain from structured light regression and detection
to image reconstruction/generation techniques.
∠ Responsible for and supervised the complete end to end development the models and system
implementation.
Implementation, Development, Deep learning, Machine learning, Python
2019 - 2020
project
SYNTHETIC REAL OBJECT AND IMAGE GENERATION
WheelRight.
The network utilises a Generative Adversarial architecture based of the ground breaking DCGAN and an STN (Synthetic Transforming Network) to synthetically generate data for object detection training. This architecture allows the training of detection & classifications networks with upwards of 50% less data to a comparable accuracy.
Deep learning, Computer vision, Python, Data Science, DATA ANALYSIS AND VISUALIZATION
2018 - 2019
project
RAPID OBJECT DETECTION AND SEGMENTATION
WheelRight.
The implementation focused on a low computationally intensive model (running at a frequency of approximately 10 Hz on a multi-thread i5 CPU) with upwards of 90% Test on site accuracy.
Computer vision, Deep learning, Data Science, Data processing, Python
2018 - 2018
project
DISSERTATION: LARGE SCALE MULTIPLE HUMAN MOTION OUTLIER IDENTIFICATION
Dissertation focuses on using large scale tracked human motion from a fixed surveillance camera in a pedestrian area, to predict and identify abnormal features of the environment as well as the motion within the environment. The two stage approach consisted of utilizing Gaussian Mixture Models & K-NN clustering to enable an environment-based motion likelihood map to be constructed. The second stage consists of motion extrapolation and prediction utilizing a CNN-LSTM architecture to sample the motion likelihood map.
University of Bristol
Deep learning, Python, Unsupervized learning, Data Analysis, Reporting
2018 - 2018
project
DEVELOPMENT OF PARTICLE FILTER & PATH PLANNING FOR ROBOTIC SYSTEMS
Focused on localisation and mapping of a robotic systems environment as well as optimal path planning with obstacle avoidance.
University of Bristol
Matlab, Robotic process automation, Automation
2015 - 2015
internship
SIGNAL PROCESSING ENGINEER | INTERNSHIP
Telumat, ROBOTICS.
November 2014 ∠ Interned in the UAV RND sector of the company.
∠ Data handling and decoding of telematic data.
∠ Interfacing data with a UI on the systems bay station.
∠ Data handling and decoding of telematic data.
∠ Interfacing data with a UI on the systems bay station.
UI, Signal processing, Processing, Internship
My education
2017
-
2018
University of Bristol
First Class Honours, Robotics
First Class Honours, Robotics
2013
-
2016
University of Cape Town
First Class Honours, Mechanical Engineering
First Class Honours, Mechanical Engineering
Alexander's reviews
Alexander has not received any reviews on Worksome.
Contact Alexander Spanellis
Worksome removes the expensive intermediaries and gives you direct contact with relevant talent.
Create a login and get the opportunity to write to Alexander directly in Worksome.
38100+ qualified freelancers
are ready to help you
Tell us what you need help with
and get specific bids from skilled talent in Denmark