EL MEHDI HICHAM
Computer Vision Engineer | Sports AI & Real-Time Video
New to Contra
EL MEHDI is ready for their next project!
TactiVision STEP 3 — Football AI Dashboard: KPIs, Heatmaps, Pass Maps & Tactical Reports
STEP_3 (https://www.linkedin.com/search/results/all/?keywords=%23step_3&origin=HASH_TAG_FROM_FEED): Data Extraction & Analytics Engine ✅ After completing detection, tracking, re-identification, and tactical mapping, the next milestone was turning the visual pipeline into a real football intelligence dashboard. STEP 3 is now focused on transforming raw match video into structured KPIs, tactical maps, live match insights, and detailed automated reports. For this demo, I tested the system on hashtag#Morocco (https://www.linkedin.com/search/results/all/?keywords=%23morocco&origin=HASH_TAG_FROM_FEED) vs hashtag#Haiti (https://www.linkedin.com/search/results/all/?keywords=%23haiti&origin=HASH_TAG_FROM_FEED) match footage, with a full tactical dashboard running around the video feed. Special mention to Fédération Royale Marocaine de Football (https://www.linkedin.com/company/f%C3%A9d%C3%A9ration-royale-marocaine-de-football/) and FIFA World Cup 2026™ - Canada, Mexico and the United States (https://www.linkedin.com/company/fifa-world-cup-2026-united-north-america-canada-usa-mexico/)— this demo is part of my ongoing work to explore how computer vision can transform football video into tactical intelligence
1
1
156
TactiVision automatically confirms Morocco’s third goal against Haiti using a Computer Vision pipeline combining: pitch and goal keypoints YOLO26s goal-mask segmentation ball detection and tracking goal-line reconstruction temporal confirmation across consecutive frames The system first reconstructs the visible goal geometry, then evaluates the tracked ball’s position relative to the goal line and goal-mouth polygon. This is part of a broader automatic football-event understanding pipeline currently being developed for TactiVision. Morocco’s third goal in the match was scored by Soufiane Rahimi in the 78th minute.
#TactiVision (https://www.youtube.com/hashtag/tactivision) #Morocco (https://www.youtube.com/hashtag/morocco) #FootballAI (https://www.youtube.com/hashtag/footballai) #ComputerVision (https://www.youtube.com/hashtag/computervision) #SportsAnalytics (https://www.youtube.com/hashtag/sportsanalytics)
1
173
This TactiVision demo analyses Harry Kane’s penalty against Paris Saint-Germain using an automatic goal-confirmation pipeline. The system combines: goal keypoint estimation confidence-aware filtering goal-line geometry YOLO26s goal-mask segmentation ball detection and temporal tracking multi-frame goal confirmation Unlike standard object detection, the objective is not simply to detect the goal or the ball. The system must determine whether the ball crosses the reconstructed line inside the valid goal-mouth region. Kane scored the penalty in the 17th minute of Bayern’s Champions League semi-final first leg against PSG.
#HarryKane (https://www.youtube.com/hashtag/harrykane) #BayernMunich (https://www.youtube.com/hashtag/bayernmunich) #PSG (https://www.youtube.com/hashtag/psg) #ChampionsLeague (https://www.youtube.com/hashtag/championsleague) #FootballAI (https://www.youtube.com/hashtag/footballai) #TactiVision (https://www.youtube.com/hashtag/tactivision)
1
176
Detecting a goal is easy. Knowing whether the ball actually crossed the line is a geometry problem.
In this #TactiVision R&D module, I combine:
goal keypoint detection to anchor the goal geometry
goal segmentation to estimate the goal area
confidence filtering
geometric projection
polygon-based validation
temporal stabilization The main challenge is robustness: a single weak keypoint near the post can shift the projected goal line, so the system must decide which geometric information can be trusted before confirming a goal. This module is part of my broader football Computer Vision pipeline for detection, tracking, pitch mapping and tactical analysis
1
178