The dataset contains BioCam visual seafloor mapping device from data collected between 23rd September to 5th October 2022. These data were collected by the University of Southampton and the National Oceanography Centre (NOC) as part of the INSITE (Influence of man-made structures in the ecosystem) AT-SEA (Autonomous Techniques for anthropogenic Structure Ecological Assessment NE/T010649/1) project. Two shore-launched Autonomous Underwater Vehicles (AUVs) deployments were conducted in the North Sea, at the site of the decommissioned North West Hutton oil platform and Miller platform. These data include colour corrected strobed images, and cm-resolution bathymetry maps and texture maps. These data were collected using the BioCam seafloor mapping device mounted to the 6000 m rated Autosub Long Range (ALR). To collect colour imagery, a strobe was mounted at the front and another one at the back of the Autonomous Underwater Vehicle (AUV) and were used to illuminate the seafloor when the colour camera of BioCam, mounted at the centre of the AUV, acquired those images once every 3 s. The strobed colour images were stored in raw format along with their timestamps. A line laser mounted at the front and another one mounted at the back of the AUV projected lines onto the seafloor at the same time. The lasers were permanently on, except when the strobes were triggered, when they were briefly turned off to avoid projecting the laser lines onto the strobed colour photos. Images of the laser line projection were acquired at 10 Hz and saved along with their timestamps. Post mission, the strobed images were colour corrected with an algorithm implemented in oplab-pipeline in post processing. Bathymetric data were computed using the laser line images that were processed with a light-sectioning algorithm published by Bodenmann, Thornton and Ura (2016). Texture maps were generated by projecting the colour-corrected images onto the 3D reconstructed bathymetry as detailed by Bodenmann, Thornton and Ura (2016).