Video data collected from both sides (NNW → SSE and SSE → NNW directions) of a row of Jaen (VIVC-5652) grape variety, trained in a VSP system, under natural canopy and lighting conditions. A Insta360 X4 camera (see https://www.insta360.com/product/insta360-x4), mounted on an extendable selfie-stick support, was operated manually along the vineyard rows, covering approximately 85 m per row.The recordings were captured in single-lens mode at 30 frames per second and 4K resolution (3840×2160), providing a 180° FOV. The continuous video was converted into 147 individual images by extracting a frame every three seconds to avoid repetitiveness between images.All images were manually annotated and each visible inflorescence was enclosed within a rectangular bounding box toidentify its the position, scale, and class, standardized to the YOLO format.

This work was co-financed by "Component 5: Capitalization and Business Innovation",integrated in the Resilience Dimension of the Recovery and Resilience Plan within the scope of the Recovery and Resilience Mechanism (MRR) of the European Union (EU), framed in the NextGenerationEU, for the period 2021–2026 within the project “Wine4cast – Space-time prediction of wine productivity for multi-actor usability: integration of remote optical-photonic sensors, artificial intelligence and climate scenarios” (ref: PRR-C05-i03-I-000071) (https://www.fc.up.pt/Wine4Cast/Home.html).