Share

Two researchers from the University of Cape Town’s (UCT) Faculty of Engineering & the Built Environment (EBE) have become the institution’s first-ever recipients of search engine giant Google’s Research Scholar Program.

The two recipients are Associate Professor Amir Patel and Dr Mohohlo Tsoeu, both from the Department of Electrical Engineering and its newly formed African Robotics Unit. They are also the only Africans in the programme’s 2021 cohort.

The programme aims to support early-career researchers who are pursuing research in fields relevant to Google. It provides “unrestricted gifts to support research at institutions … and is focused on funding world-class research”.

To be eligible for the programme, an entrant’s research must be in computer science and related fields. Associate Professor Patel has been accepted in the Machine Perception category and Dr Tsoeu in the Natural Language Processing category.

“It feels amazing to be recognised by Google, one of the largest tech companies in the world. It is also exciting and encouraging for us to be mentioned among some of the world’s top universities in the field of robotics and artificial intelligence,” said Patel.

Dr Mohohlo Tsoeu is participating in the Natural Language Processing category. Photo Je’nine May.

Tsoeu said that he was “excited and honoured” to be one of only two researchers from an African university to receive this prestigious honour this year.

“It is a positive affirmation that our research is important and has the potential to have great impact, and that we have the intellectual capital to deliver,” he said.

Invaluable, unprecedented insight

Patel was awarded for his research project, “WildPose: 3D animal biomechanics in the field using multi-sensor data fusion”.

The project aims to provide deeper insight into the abilities of the world’s greatest animal athletes (located in Africa), such as the cheetah, and into how they can “robustly traverse through the unstructured world”; it will prove invaluable for legged robots if they are ever to leave the confines of the laboratory.

Patel explained that currently the biggest impediment to a holistic understanding of animal locomotion is measuring and modelling whole-body motion in the wild. Therefore, his project proposes developing a deep learning-based motion-capture system (WildPose), which leverages complementary sensor data to remotely obtain high-speed, whole-body 3D animal kinematics in the field from a single view.

The WildPose system will enable videographers to capture biomechanical data from animals such as cheetahs and lions in the wild, using a single handheld device, and creating a new source for data collection.

“This research is important, as it will allow us to measure the motion of animals in the wild at an unprecedented level,” said Patel.

Assoc Prof Amir Patel’s research will enable the measurement of animal motion in the wild at an unprecedented level. Photo Supplied.

 

Furthermore, Patel believes the proposed system will be disruptive in the field of ecology by providing new insight into how climate change affects animal behaviour; in neuroscience (how animals such as the cheetah respond to the motion of prey during hunts); in evolutionary biology (how the cheetah – the only surviving species in the Acinonyx genus – evolved, compared to Africa’s other big cats); and in robotics (holistically understanding animal locomotion will enable the design of more agile robots).

On the importance of Google’s Research Scholar Program and how it will support his research, Patel said that it is vital for enabling emerging researchers to explore artificial intelligence and robotics in exciting new domains.

“I believe this award will help me further my goal of moving biomechanics beyond the confines of the laboratory,” he said.

Restoring dignity

Tsoeu was awarded for his project, “Corpora collection and complete natural language processing of isiXhosa, Sesotho and South African Sign Language”.

His project will contribute to the development of comprehensive, high-quality language corpora for indigenous South African languages. It will also investigate and develop novel and high-performance machine-learning algorithms aimed at application areas such as automatic speech recognition, translation and text-to-speech/sign technology.

Dr Mohohlo Tsoeu hopes to bridge divides and restore dignity through his research. Photo Je’nine May.

 

Tsoeu explained that these applications are in the growing area of human–machine interfacing. But more importantly in the South African context, they contribute towards bridging the human language divide and improving equal access and participation to restore the dignity of currently marginalised groups, such as the Deaf and hard-of-hearing communities.

“My passion is to do research that drives good social change,” he said.

“I believe that in conducting such research, one opens avenues for good fundamental and applied research questions that contribute new knowledge and advance scholarship.”

His research interests are in machine learning, with specific applications to aspects of natural language processing. This includes language designing, language corpora for machine learning, investigating novel algorithms for language translation, automatic speech and sign recognition, and text-to-speech and sign synthesis. Tsoeu said that it is a multi-disciplinary research area that stimulates collaborations between engineering, linguistics and the psychology of learning, and as he puts it, “is really fascinating”.

Bridging the divide

Tsoeu’s research contributes to “bridging the human-to-human language divide that devastates South Africa, leading to political language debates at universities and other spaces, and marginalisation of native speakers of native languages, especially the Deaf community”.

“The world is getting extremely connected, both through travel and the web, and the language divide remains a bottleneck to enjoying full global connectedness,” said Tsoeu.

In addition to contributing to new knowledge and new language corpora, Tsoeu’s and his team’s work will also open doors to “amazing research ideas and innovation in the areas of machine learning, human–machine interfacing, media content distribution, and education”.

But all of this costs money, and a lot of it. Therefore, the funding Tsoeu will receive from the programme will make a significant contribution to the project’s “enormous costs”, which include his and his team’s research activities, student funding, travel costs, fieldwork and attending conferences.

“I am honoured to have received it,” said Tsoeu.

 

APRIL 2021 | STORY CARLA BERNARDO