Connelly and Clandinin (1986) point out that the distinction between reflection-in-action and reflection-on-action “…separates thinking during practice from thinking after or before” but finds that “…both terms name the method used in the act of thinking practically.” (p.294) The following reflections are undertaken following completion of the project and begin with a review of the experience. As a practitioner I found that approaching this project was different to any I had undertaken previously. I found I was constantly referring to the methodical instruments defined in my methodology, which Initially proved burdensome and had a retarding effect on my progress. However, over time I became more familiar with the tools, the ways of thinking and both the methods and frequency of documenting, and did find the process became less comfortable, akin to a ‘new normal’.
Perhaps the greatest challenge, and one which continued through the entire project lifecycle, came from the incongruous act of attempting to externalise the tacitly understood insights by asserting knowledge and specific skills that I knew, and differentiating from those that I didn’t. I found my online journal served me well in terms of providing an outlet for recording my actions and observations without the constraints and disciplines associated with academic writing. It has also provided an excellent repository for the diagrams, images, audio visual recording and renders that have been developed along the lifecycle of the project. My regret is that the online journal is yet to gain any meaningful traction and therefore not yet providing the backtalk between academics and practitioners that I hoped it would.
At the outset I intended to use diagnostic sketches in situations where a problem had emerged but with multiple potential options for resolution. This format allows the various implications associated with each to be quickly mapped out. Whilst I barely needed to use these on this project, I maintain that the problems presented by upcoming projects in the case study will be more complex and a means of quickly laying out the options will be needed.
I anticipated that audio-visual screen recordings would be an important means of articulating and reflecting on my approach and this proved to be correct. The nature of undertaking a PhD on a part-time basis require the project to sit dormant for periods of time, but I found that skimming through the online journal, and running the audio-visuals allowed me to quickly reacquaint, at least in relation to the practice-based work. I used these more than any other tool for seeking out tacit knowledge and subliminal bias.
In addition to reviewing the process from an experiential perspective, I also intended that reflection-on-action would allow me to draw insights from the project to establish new areas of exploration on future projects. In this case I believe that the proposition was reasonable as a means of proving the two fundamental questions around the geometry and methods of applying textures but did not address how camera projection would respond to more expansive camera moves in which there may be loss of coverage, smearing of projected textures due to perspective distortion and loss of resolution. More complex environments and closer views would be needed to examine further the level of detail needed in the geometry, and perhaps identify the threshold at which 3D library assets would need to be discarded in favour of more fully modelled surfaces. Whilst this project served its purpose in terms of its localised objectives, it was applied to a relatively simple landscape and, whilst there were geometric ‘planar’ surfaces (i.e. church facades) in the scene, these were too small in the frame to reveal any texture smearing arising from lateral movement of the shot camera.
Whilst mapping the textures directly to the UV’s proved cumbersome in this project, I refer to insights from an online training course by Smith (2011) for FXPHD. The project entailed digital clean-up work on an environment using cameras to project the textures onto geometric objects but then UV renders were taken of the sections of the image requiring paint work. Because this creates an orthographic representation of the area, Smith demonstrated advantages to this approach when undertaking texture cloning and may therefore be beneficial on more complex projects where patch projection is required.
I am keen to undertake some academic writing and remain confident that there are sufficient areas of complexity in my topic to interest the academic community. I do however feel this should follow the second project in this case study where the technical challenges will be greater and wider utilisation of the methodological instruments.
In terms of dissemination and presentation of content, the material generated from this case study has already been used in teaching. Some of the technical aspects have been converted into learning materials for visual effects students and the academic aspects of the broader study has been presented to post-graduate students as part of a mini research seminar.
In summary I refer to the objectives that were developed for this project and the findings arising from the practical work.
The concepts surrounding application of the matte painting (or sections of) were thoroughly investigated and clearly demonstrate that the ‘camera projection’ method is, in almost all cases, advantageous. This finding perhaps carries the greatest importance as it serves as a proof of concept, which can be deployed across future projects in this study and negating any significant further explanation or discussion.
The topic of geometry and its use as a surface for projecting sections of a matte painting is also proven in the sense that the topology should be as light and simple as possible. However the project demonstrated that this has to be considered in isolation for every projection. Of greater significance was the alignment of the geometry to visual landmarks, or the ‘perceived’ topology, in the artwork. Incorrect execution correlated with surfaces failing to display parallax correctly and the effect of textures sliding and separating, due to poor surface contact.
These finding must however be considered in the context of the design of this project and in particular the relatively benign camera movement. Whilst camera projection was proven to be advantageous, it was possible to texture the surface from a single projection camera without loss of coverage or resolution and without any stretching or smearing of the textures across the various surfaces. Future projects will need to be designed with more expansive camera animation, both in terms of lateral (tracking/trucking) movement and much more dynamic forward or backward (dollying) movement. https://vimeo.com/manage/videos/522924471
Peter Holmes PhD Journal 