Intelligent Video Processing Algorithm Engineer

PERSOL SINGAPORE PTE. LTD.; Singapore; On-site; SGD 70,000 - 110,000; Full time

Job summary

PERSOL SINGAPORE PTE. LTD. is seeking a highly skilled researcher to advance intelligent video processing algorithms for multimedia and communication applications in Singapore.

The role involves exploring computer vision, pattern recognition, and machine learning methods, with emphasis on CNN/transformer models, feature extraction, and real-time video analysis. Collaboration with hardware teams for DSP implementations is expected.

Qualifications

• Master’s degree or above in CS/Robotics/EE with focus on computer vision or robotics perception.
• Strong knowledge of deep learning, CNNs/Transformers, and traditional ML for recognition.
• Experience in PC-based algorithm development for machine vision projects.
• Experience implementing low-level vision algorithms on embedded DSP platforms.
• Proficient in C/C++ and MATLAB; familiar with OpenCV.
• Familiar with model compression techniques such as pruning or distillation.
• Experience with PyTorch and TensorFlow.
• Publications at CVPR/ICCV/NIPS/TPAMI are a plus.
• Experience with large CV and language models is advantageous.

Responsibilities

• Participate in research of advanced intelligent video processing algorithms using state-of-the-art recognition technologies for multimedia and communication applications
• Research in intelligent video processing, computer vision, pattern recognition or machine learning for applications like object detection/tracking, event recognition, video analysis
• Perform research into leading edge and emerging video technologies, ideas and approaches and keep up to date with new video processing standards and general industry trends
• C-model simulation for video processing algorithm development and verification.

Skills

Deep learning

CNN/Transformer

Computer vision

Pattern recognition

Visual tracking

Feature extraction

Pytorch

TensorFlow

OpenCV

C/C++

Matlab

Model compression

Research communication

Education

Master’s degree or above in Computer Science, Robotics, Electrical Engineering, or related field with a focus on computer graphic, computer vision or robotics perception

Tools

OpenCV

TensorFlow

PyTorch

DSP platforms

MATLAB

Job description

Key Responsibilities:

• Participate in research of advanced intelligent video processing algorithms using state-of-the-art recognition technologies for multimedia and communication applications
• Research in intelligent video processing, computer vision, pattern recognition or machine learning for applications like object detection/tracking, event recognition, video analysis, etc.
• Perform research into leading edge and emerging video technologies, ideas, approaches and keep up to date with new video processing standards and general industry trends
• C-model simulation for video processing algorithm development and verification.

Required Qualifications:

• Master’s degree or above in Computer Science, Robotics, Electrical Engineering, or related field with a focus on computer graphic, computer vision or robotics perception.
• Insightful know-how on deep-learning, CNN/transformer, traditional machine learning and recognition technologies, such as human and object feature extractions, visual tracking, event recognition
• PC-based algorithmic intelligent video processing algorithm development for machine vision projects and applications.
• Implementation of low-level vision algorithms on embedded DSP platforms with associated CPU and memory bandwidth limitations
• Working knowledge of C/C++ and/or Matlab and proficiency with DSP and filter design.
• Familiarity with ASIC design approach and environment Familiar with OpenCV library and related algorithms
• Ability to express ideas and communicate effectively.
• Insightful understanding of compact CNN model and design, such as ShuffleNet, MobileNet, SqueezeNet, etc.
• Be familiar with model compression algorithms on video processing, such as knowledge distillation, parameter pruning and sharing, low-rank factorization, etc.
• Rich experiences in handling the deep learning development tools, such as Pytorch, TensorFlow, etc.
• Any of the following skills and experiences would be an added advantage:
• Key image processing algorithms such as super resolution/HDR/contrast enhancement for display devices.
• Publications on top-ranking conferences and journals in related fields, such as CVPR, ICCV, NIPS, TPAMI, etc.
• Top-ranking and/or award receivers in world competition in video processing field
• Research experience on large cv/language model