此崗位根據(jù)業(yè)務(wù)需求,不定期出差法國,不能接受勿擾?。。?/div>
一、工作任務(wù):
1、開發(fā)和優(yōu)化計算機視覺算法,使機器人能夠感知、理解周圍環(huán)境并與之交互。
2、將視覺系統(tǒng)與人工智能、運動控制和云服務(wù)相集成,以實現(xiàn)強大的實時性能。
3、與硬件、人工智能和嵌入式團隊合作,確保視覺模型在嵌入式或邊緣設(shè)備上高效部署。
4、通過數(shù)據(jù)驅(qū)動的開發(fā)、測試和調(diào)優(yōu),持續(xù)提升感知能力。
二、主要職責(zé):
1、設(shè)計和實現(xiàn)用于目標(biāo)檢測、識別、跟蹤和場景理解的算法。
2、利用 RGB、深度傳感器、激光雷達和多相機設(shè)置開發(fā)二維和三維視覺處理流程。
3、優(yōu)化視覺模型,以在嵌入式平臺(GPU、TPU、VPU、DSP)上實現(xiàn)實時推理。
4、構(gòu)建用于訓(xùn)練和驗證視覺模型的數(shù)據(jù)收集和標(biāo)注流程。
5、將計算機視覺模塊與機器人軟件棧(ROS/ROS2 或自定義框架)相集成。
6、確保在復(fù)雜環(huán)境條件下(光線變化、遮擋、背景移動等)的穩(wěn)定性。
7、與人工智能 / 機器學(xué)習(xí)工程師合作,在生產(chǎn)環(huán)境中部署和維護模型。
8、在真實機器人平臺和仿真環(huán)境中測試和驗證感知功能。
9、維護文檔、保證代碼質(zhì)量并進行版本控制,以確保長期可維護性。
三、所需技能:
專業(yè)技術(shù)技能:
1、具備扎實的 C++ 和 Python 使用經(jīng)驗
2、有使用計算機視覺庫(OpenCV、Open3D、PCL)的經(jīng)驗
3、了解人工智能 / 機器學(xué)習(xí)框架(PyTorch、TensorFlow、ONNX 等)
4、熟悉嵌入式系統(tǒng)(GPU、TPU、NPU 等)的模型優(yōu)化
5、有數(shù)據(jù)管理和標(biāo)注經(jīng)驗(數(shù)據(jù)集創(chuàng)建、清理、標(biāo)記)
6、理解三維數(shù)據(jù)處理、點云以及同步定位與地圖構(gòu)建(SLAM)
7、了解用于模型訓(xùn)練的數(shù)據(jù)增強和合成數(shù)據(jù)生成
軟技能:
1、具備以解決問題為導(dǎo)向的思維和注重成果交付的執(zhí)行力
2、能夠與技術(shù)和非技術(shù)人員進行良好溝通
語言能力:法語和英語達到專業(yè)熟練水平
教育背景:計算機視覺、機器人學(xué)、人工智能或相關(guān)領(lǐng)域的碩士或博士學(xué)位
工作經(jīng)驗:2 年計算機視覺開發(fā)經(jīng)驗,有機器人或自主系統(tǒng)相關(guān)經(jīng)驗者優(yōu)先
Job Title: Computer Vision Developer – Robotics Perception & AI
Mission:
1. Develop and optimize computer vision algorithms to enable robots to perceive, understand, and interact with their environment.
2. Integrate vision systems with AI, motion control, and cloud services for robust real-time performance.
3. Collaborate with hardware, AI, and embedded teams to ensure efficient deployment of vision models on embedded or edge devices.
4. Continuously improve perception capabilities through data-driven development, testing, and tuning.
Main Responsibilities:
1. Design and implement algorithms for object detection, recognition, tracking, and scene understanding.
2. Develop 2D and 3D vision pipelines using RGB, depth, LiDAR, and multi-camera setups.
3. Optimize vision models for real-time inference on embedded platforms (GPU, TPU, VPU, DSP).
4. Build data collection and annotation pipelines for training and validating vision models.
5. Integrate computer vision modules with the robot software stack (ROS/ROS2 or custom frameworks).
6. Ensure robustness in challenging conditions (variable lighting, occlusion, moving backgrounds).
7. Collaborate with AI/ML engineers to deploy and maintain models in production.
8. Test and validate perception features on real robotic platforms and in simulation.
9. Maintain documentation, code quality, and version control for long-term
maintainability.
Skills Required:
Technical:
1. Strong experience in C++ and Python
2. Experience using computer vision libraries (OpenCV, Open3D, PCL)
3. Knowledge of AI/ML frameworks (PyTorch, TensorFlow, ONNX…)
4. Familiarity with model optimization for embedded systems (GPU, TPU, NPU…)
5. Experience in data management & annotation (dataset creation, cleaning, labeling)
6. Understand 3D data processing, point clouds, and SLAM.
7. knowledge of data augmentation and synthetic data generation for training models.
Soft Skills:
1. Problem solving oriented thinking and delivery-focused execution
2. Excellent communication skills across technical and non-technical audiences
Languages:
Professional proficiency in French and English
Education:
Master’s or PhD in Computer Vision, Robotics, AI, or related fields.
Experience:
2 years in computer vision development, preferably in robotics or autonomous systems.