Assessing the Quality of Soccer Shots from Single-Camera Video with Vision-Language Models and Motion Features
Bibliographic record
Abstract
Automated quality assessment (AQA) of football striking technique typically relies on wearable inertial units or multi-camera marker-based systems that impose substantial logistical overhead. We propose a single-camera, vision-language pipeline that combines raw video with engineered 3D kinematic features and leverages a Vision-Language Model (VLM) to generate expert-style feed-back and scalar ratings. A monocular side view recording of each shot is first processed with RTMPose-3D [9] and smoothed with a 1-Euro filter [6]. Nine domain-informed features spanning the approach phase, plant mechanics, segmental coordination, and post-contact stability are automatically extracted and, together with the video clip, form a multimodal prompt to the VLM. Experiments on 70 shots (50 adults, 20 youth) labeled by UEFA B-licensed coaches demonstrate that the joint video and feature prompt achieves an MAE of 0.61 ± 0.03 with the ground truth, out-performing the video-only and feature-only variants by 0.34 and 0.24, respectively. Ablations reveal that down-sampling the frame rate to 30 FPS produces minimal quality degradation, and that 1-Euro temporal filtering slightly reduces the rating error.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".