Abstract Field-based next generation phenotyping has become of great interest to plant breeders and agricultural researchers in recent years, particularly for circumventing destructive or impractical phenotyping methods commonly used for certain traits. The non-destructive estimation of one such trait, above ground biomass (AGB), has been investigated repeatedly using 2D imagery, though little research has been conducted on 3D methods. The aims of the current study were to (i) investigate the use of readily-available consumer level digital cameras and software to estimate AGB, canopy height (CH) and harvest index (HI) of wheat plots, (ii) investigate the suitability of this data as a replacement for destructive sampling methods within a wheat breeding programme, and (iii) identify the point cloud density required for accurate estimation of AGB. To achieve this, a small plot trial of a single wheat cultivar was conducted in an irrigated nursery, at Roseworthy, South Australia. At physiological maturity plots were measured for CH and whole plots were harvested to attain AGB and threshed to measure grain yield and calculate HI. Prior to harvesting each plot was imaged using a digital camera, with these images being processed into 3D point clouds, which were subsequently used to estimate plot volume and CH. Strong correlations were observed between actual measurements of AGB, CH and HI to those estimated from point clouds. Images were processed in subset batches to determine an optimal number of images for processing. Stronger correlations between AGB and plot volume were observed when more images were processed, though as few as 48 images provided sufficiently accurate estimates of AGB. These methods were shown to be effective at estimating AGB, CH and HI and could be adopted by small scale research programmes. This study shows that a higher-throughput adaptation of this photogrammetry method could be used in phenotype intensive research such as plant breeding programmes.