Kashif M.

→ 𝟏𝟐+ 𝐲𝐞𝐚𝐫𝐬 𝐨𝐟 𝐡𝐚𝐧𝐝𝐬-𝐨𝐧 𝐞𝐱𝐩𝐞𝐫𝐢𝐞𝐧𝐜𝐞 → 𝐆𝐨𝐥𝐝 𝐌𝐞𝐝𝐚𝐥𝐢𝐬𝐭 𝐢𝐧 𝐌𝐚𝐬𝐭𝐞𝐫𝐬 𝐨𝐟 𝐃𝐚𝐭𝐚 𝐒𝐜𝐢𝐞𝐧𝐜𝐞 (2012) → Former 𝐀𝐈/𝐌𝐋 𝐏𝐫𝐨𝐟𝐞𝐬𝐬𝐨𝐫 (5 𝐲𝐞𝐚𝐫𝐬) 𝐚𝐧𝐝 𝐔𝐝𝐞𝐦𝐲 𝐈𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐨𝐫 (13 𝐩𝐮𝐛𝐥𝐢𝐬𝐡𝐞𝐝 𝐀𝐈/𝐌𝐋 𝐜𝐨𝐮𝐫𝐬𝐞𝐬) → 𝐏𝐮𝐛𝐥𝐢𝐬𝐡𝐞𝐝 𝐫𝐞𝐬𝐞𝐚𝐫𝐜𝐡 𝐩𝐚𝐩𝐞𝐫𝐬 𝐢𝐧 𝐭𝐨𝐩 𝐢𝐧𝐭𝐞𝐫𝐧𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐀𝐈/𝐌𝐋 𝐜𝐨𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞𝐬 𝐂𝐨𝐫𝐞 𝐄𝐱𝐩𝐞𝐫𝐭𝐢𝐬𝐞 𝐢𝐧 𝐂𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐕𝐢𝐬𝐢𝐨𝐧: • 𝐎𝐂𝐑 & 𝐓𝐞𝐱𝐭 𝐄𝐱𝐭𝐫𝐚𝐜𝐭𝐢𝐨𝐧: Tesseract, EasyOCR, PaddleOCR, CRNN, Invoice/ID Parsing, License Plate Recognition • 𝐎𝐛𝐣𝐞𝐜𝐭 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 & 𝐓𝐫𝐚𝐜𝐤𝐢𝐧𝐠: YOLO, Faster R-CNN, EfficientDet, Detectron2, DeepSORT, ByteTrack, SORT • 𝐒𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧: U-Net, Pix2Pix, Mask R-CNN, DeepLabv3+, SegNet, FCN, SAM, Semantic & Instance Segmentation • 𝐅𝐚𝐜𝐢𝐚𝐥 𝐀𝐧𝐚𝐥𝐲𝐬𝐢𝐬: FaceNet, Dlib, MTCNN, RetinaFace, InsightFace, Emotion Recognition, Age/Gender Prediction, Face Landmark Tracking • 𝐏𝐨𝐬𝐞 𝐄𝐬𝐭𝐢𝐦𝐚𝐭𝐢𝐨𝐧: DCPose, OpenPose, MediaPipe, PoseNet, HRNet, Skeleton Tracking, Gesture Recognition • 𝐕𝐢𝐝𝐞𝐨 𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 & 𝐒𝐮𝐫𝐯𝐞𝐢𝐥𝐥𝐚𝐧𝐜𝐞: Object Counting, Vehicle Detection, Traffic Analytics, Action Detection, Anomaly Detection • 𝐄𝐝𝐠𝐞 & 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐈 𝐃𝐞𝐩𝐥𝐨𝐲𝐦𝐞𝐧𝐭: TensorRT, ONNX Runtime, TFLite, Quantization, AIMET, Jetson Nano • 𝐈𝐦𝐚𝐠𝐞 𝐄𝐧𝐡𝐚𝐧𝐜𝐞𝐦𝐞𝐧𝐭 & 𝐑𝐞𝐬𝐭𝐨𝐫𝐚𝐭𝐢𝐨𝐧: Super-Resolution, Autoencoders, Deblurring, Inpainting, GAN-based Enhancement • 𝟑𝐃 𝐕𝐢𝐬𝐢𝐨𝐧 & 𝐃𝐞𝐩𝐭𝐡 𝐄𝐬𝐭𝐢𝐦𝐚𝐭𝐢𝐨𝐧: Stereo Vision, Monocular Depth Prediction, Hand & Object 3D Pose Estimation • 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐕𝐢𝐬𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬: Pix2Pix, CycleGAN, Stable Diffusion, ControlNet, Synthetic Data Generation ⸻ 𝐓𝐨𝐨𝐥𝐬, 𝐅𝐫𝐚𝐦𝐞𝐰𝐨𝐫𝐤𝐬 & 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦𝐬 • 𝐅𝐫𝐚𝐦𝐞𝐰𝐨𝐫𝐤𝐬: PyTorch, TensorFlow, Keras, Detectron2, OpenMMLab, Scikit-Learn, MMDetection, Ultralytics YOLO, FastAI • 𝐂𝐕 𝐋𝐢𝐛𝐫𝐚𝐫𝐢𝐞𝐬: OpenCV, Pillow, Albumentations, Mediapipe, PyTorch Lightning, ImgAug • 𝐎𝐩𝐭𝐢𝐦𝐢𝐳𝐚𝐭𝐢𝐨𝐧 & 𝐌𝐋𝐎𝐩𝐬: Weights & Biases, MLflow, ClearML, Optuna, TensorBoard, Docker, DVC, ONNX, TensorRT, CUDA • 𝐃𝐞𝐩𝐥𝐨𝐲𝐦𝐞𝐧𝐭 & 𝐀𝐏𝐈𝐬: FastAPI, Flask, TorchServe, Triton, AWS Lambda, Azure ML, GCP Vertex AI • 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞𝐬: Python, C++, C#, JavaScript, SQL, Bash • 𝐕𝐢𝐬𝐮𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧 & 𝐃𝐚𝐭𝐚 𝐓𝐨𝐨𝐥𝐬: NumPy, Pandas, Matplotlib, Seaborn, Plotly • 𝐏𝐥𝐚𝐭𝐟𝐨𝐫𝐦𝐬: AWS, Azure, GCP, NVIDIA Jetson, Qualcomm AI Hub, Raspberry Pi, Edge TPU ⸻ 𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐝 𝐂𝐨𝐦𝐩𝐮𝐭𝐞𝐫 𝐕𝐢𝐬𝐢𝐨𝐧 𝐏𝐫𝐨𝐣𝐞𝐜𝐭𝐬 Sports Pose & Equipment Tracking | DCPose + YOLO Created a tennis analytics model tracking player poses, racket, and ball trajectories for motion feedback. Shoplifting Detection System | Object Detection & Action Recognition Built a real-time system using YOLOv8 to detect suspicious movements and theft-like actions in CCTV feeds. Face Landmark Detection for Snapchat-Style Filters | Facial Landmark Tracking Integrated MediaPipe-based face landmark detection for instant AR filter placement in mobile apps. License Plate & Car Make Recognition | OCR + Object Detection Used YOLO and EasyOCR for license and car make recognition for parking and traffic automation. VisionAssist for the Visually Impaired | Object Detection + Distance Estimation Developed a YOLOX-Tiny Android app that announces nearby objects with distance and direction. Traffic Analytics using YOLOX | Object Detection & Segmentation Implemented vehicle detection and counting for congestion and road usage analytics. Hand Gesture Recognition | Pose & Contour Detection Built gesture recognition using contour and keypoint tracking for touchless control. Age & Gender Detection from Image | Facial Recognition Trained CNN-based model to predict age and gender from facial images in real time. Image to Sketch Converter | Laplacian Filter in Computer Vision Applied Laplacian filters to convert any image into a pencil-style sketch. Satellite Image Segmentation | Pix2Pix Model (GAN) Trained Pix2Pix GAN to convert satellite images into segmented maps showing roads and vegetation. ⸻ 𝐓𝐞𝐚𝐜𝐡𝐢𝐧𝐠 & 𝐑𝐞𝐬𝐞𝐚𝐫𝐜𝐡 • Taught Computer Vision, Deep Learning, and Applied AI courses at the university level. • Published papers on object detection, transfer learning, and image classification. • Created 13 AI/ML courses on Udemy (e.g., Computer Vision in Python for Beginners). ⸻ 𝐈𝐟 𝐲𝐨𝐮 𝐚𝐫𝐞 𝐢𝐧𝐭𝐞𝐫𝐞𝐬𝐭𝐞𝐝 𝐢𝐧 𝐜𝐨𝐦𝐩𝐮𝐭𝐞𝐫-𝐯𝐢𝐬𝐢𝐨𝐧 𝐩𝐫𝐨𝐣𝐞𝐜𝐭𝐬, 𝐥𝐞𝐭’𝐬 𝐡𝐚𝐯𝐞 𝐚 𝐪𝐮𝐢𝐜𝐤 𝐜𝐡𝐚𝐭/𝐦𝐞𝐞𝐭𝐢𝐧𝐠 𝐚𝐧𝐝 𝐝𝐢𝐬𝐜𝐮𝐬𝐬 𝐭𝐡𝐞 𝐬𝐨𝐥𝐮𝐭𝐢𝐨𝐧 𝐭𝐨 𝐲𝐨𝐮𝐫 𝐩𝐫𝐨𝐛𝐥𝐞𝐦.