Extended OpenTT Games Dataset: A table tennis dataset for fine-grained shot type and point outcome

TL;DR

This work extends the public OpenTTGames dataset with frame-accurate, fine-grained annotations for stroke type (left/right, forehand/backhand, and subtypes), player posture (body lean) and leg position, and rally-end outcomes. A compact annotation scheme and code-assisted labeling enable reproducible, task-relevant supervision that moves beyond simple event spotting toward tactical understanding of table tennis play. The extension preserves licensing (CC BY-NC-SA 4.0) to facilitate open benchmarking and reuse, and provides practical data summaries, a structured labeling pipeline, and baseline guidance. Limitations include a single static viewpoint and concise posture labels, with future work offering richer pose data, multi-view recordings, audio, and inter-rally context. Overall, the dataset enables more detailed analysis and benchmarking for fine-grained shot-type and point-outcome modeling in racket sports.

Abstract

Automatically detecting and classifying strokes in table tennis video can streamline training workflows, enrich broadcast overlays, and enable fine-grained performance analytics. For this to be possible, annotated video data of table tennis is needed. We extend the public OpenTTGames dataset with highly detailed, frame-accurate shot type annotations (forehand, backhand with subtypes), player posture labels (body lean and leg stance), and rally outcome tags at point end. OpenTTGames is a set of recordings from the side of the table with official labels for bounces, when the ball is above the net, or hitting the net. The dataset already contains ball coordinates near events, which are either "bounce", "net", or "empty_event" in the original OpenTTGames dataset, and semantic masks (humans, table, scoreboard). Our extension adds the types of stroke to the events and a per-player taxonomy so models can move beyond event spotting toward tactical understanding (e.g., whether a stroke is likely to win the point or set up an advantage). We provide a compact coding scheme and code-assisted labeling procedure to support reproducible annotations and baselines for fine-grained stroke understanding in racket sports. This fills a practical gap in the community, where many prior video resources are either not publicly released or carry restrictive/unclear licenses that hinder reuse and benchmarking. Our annotations are released under the same CC BY-NC-SA 4.0 license as OpenTTGames, allowing free non-commercial use, modification, and redistribution, with appropriate attribution.

Figures (5)

• Figure 1: Illustration of the frame interval containing a rally-ending event. The rally-ending will, by definition, occur within these intervals, which substantially reduces the number of frames that need to be inspected.
• Figure 2: Examples of stroke types (top row) and lean types (bottom row).
• Figure 3: A stroke in which the player is nearly fully outside the camera frame. The stroke is annotated as "right_backhand_loop unknown unknown".
• Figure 4: Examples of non-trivial foot-position configurations together with their counter-examples, demonstrating how the foot-position classification rules are applied.
• Figure 5: Representative examples of the rally-ending categories defined in the dataset.