The advancement in imaging technologies with computer vision-based methods have facilitated non-invasive plant trait analysis for Precision Agriculture. These traits are primarily derived from leaf level analysis of plant images, underlining the importance of leaf instance segmentation and counting tasks (termed as leaf phenotyping). To advance the development of state-of-the-art methods for the aforementioned tasks, various plant datasets have been proposed. However, these datasets comprises of model plants with uniform leaf structures. This limits the applicability of these methods on classical plants such as rice and wheat, which exhibit variability in leaf shape, size, and arrangement. To address this bottleneck, we introduced a novel dataset comprising of high-resolution rice and wheat plant images, annotated at leaf instance level. Based on this dataset, the competition “ICPR 2024 Leaf Inspect” addressed computer vision challenges in: (a) Leaf instance segmentation and (b) Leaf counting tasks. This paper report and discuss methods and findings of the participating teams. The proposed benchmark dataset will facilitate computer vision research on non-rigid objects with high degree of self-similarity and self-occlusions (Leaf Inspect Competition Website: https://sites.google.com/view/icpr-2024/ .).