Is Nerf The End of Photogrammetry

photography has been the go-to solution in the 3D World for quite some time by allowing users to turn ordinary photographs into beautiful 3D models and while it's been a lifesaver for many Industries like game development VFX and arcviz let's face it it's not without issues from waiting for hours for the process to complete to getting awful 3D scans in some cases which make it sometimes a real headache but there is a new technology that is about to shake things up and it is called Nerf this revolutionary technology has the potential to transform 3D scanning as we know it by creating stunning 3D models faster and more accurately compared to ever before so in today's video we're going to explore this brand new technology and what makes it different compared to traditional 3D scanning before we continue if you are interested in learning more about 3D and photogrammetry I recommend you try skillshare you might know skillshare for photography video editing and illustration classes but it actually has many animation game development and VFX focused stuff for example Smith sculpt on skillshare or just Smith on YouTube created this great course on photogrammetry using only your phone and reality capture this class is aimed at anyone and everyone whether you are experienced or just you want to get your feet wet with the first time using only a phone and some free software and the great thing is you can access all of them great classes all with the price of less than one course every month let alone a ton of other classes at your disposal so the first 1000 people to click the link in the description will get a one month free trial of skillshare Premium membership Nerf generates a flexible and dynamic structure called a Radiance field by utilizing the power of machine learning and just like photogrammetry this enables us to generate for realistic 3D representations from images and also videos this Radiance field allows us to see a render of objects from any angle even those that were not captured in the original input data as it guesses how the missing viewpoints will look like and this is what scary and exciting about it North and photography offer different pros and cons one of the main benefits of Nerf is its efficiency and speed as it is faster and a more intelligent technique so to speak it can create a 3D model from just few images compared to photogrammetry which requires a large number of high quality photos from the biggest amount of angles additionally Nerf can capture Dynamic objects in Motion in other terms animated models or traditional fability scanning is limited to static object this is because Nerf can use videos as input as we previously mentioned however photography remains widely used and understood in various Industries plus it offers more Universal file formats making it a tried and tested alternative for those who prefer a more established approach Nerf takes a unique approach to construct scenes but first of all we start the process in a similar way to photogrammetry since the first step involves capturing images or videos of the desired object while it doesn't have to be the same amount of photos compared to photogrammetry it never hurts to have extra data once your software starts processing the data to predict or let's say to produce a volumetric representation of the scene it uses what is known as neural network a machine learning algorithm that is trained to generate the radiance field of the scene from the provided data while guessing and filling the missing viewpoints the process is much more complicated than that but this more or less can sum it up this allows for highly realistic renders that can be viewed from any angle in a sense a Radiance field is an image or a video that could be viewed from all angles one other than traditional 3D geometry making the implementation of data a bit harder for traditional 3D software currently nevertheless there are ways to convert the data and with future improvements the expectations are high for more efficient workflows when it comes to Quality both nerve and photogrammetry have their strengths and weaknesses generally speaking Nerf produces highly realistic and DTM 3D renders so especially for objects with complex geometry and appearance because nerve can capture subtle details and it can do it better with fewer data points as well as the ability to generate reflective 3D models which is something photogrammetry can do right now in addition to the ability to render and change the light on top of having the ability to fill the missing viewpoints all of which are elements that makes North more pleasant to work with as well as being able to generate animated objects on the other hand photogrammetry is generally better suited for real-time Productions since anything that needs geometry based data has to go through photogrammetry at the moment just because it seems more efficient in my opinion rather than generating Nerf data and then being converted into a 3D mesh and despite not being as detailed or as smart it is still a great solution that can give you a good result and qixels mega scans are the best proof of that from a VFX standpoint Nerf is a game changer that holds enormous potential in the industry the final result is all that matters and Nerf's ability to create highly realistic 3D models with limited data is a significant Breakthrough by leveraging Nerf VFX artists can generate stunning 3D models with the speed and a degree of accuracy and detail that was previously impossible which they can then use to create worlds or backgrounds or even hero assets it also enables us to integrate real-life filming locations into the digital realm which makes the possibilities of creative expansions endless because it's much easier to integrate CGI that way especially when we add some nerves features such as re-lighting as well as making tasks like tracking much more pleasant in general at least in theory and even though the current motion technology is limited imagine the possibilities of capturing the entire scene of actors performing in 3D the idea itself is mind-blowing but only time will tell how effective it will be in the world of gaming however integrating new technologies can be a challenge when it comes to Nerf the main obstacle is the fact that the game engine is predominantly used 3D mesh which are based on triangular faces and Nerf being a different type of data makes it quite hard to integrate seamlessly while there are ways to convert this data the process of converting a Nerf model into geometry involve extracting the positional data from the Nerf model and usually need to create a 3D mesh while this method can be effective it is not without its limitations meaning you have to fix and optimize the generator geometry which is anything other than straightforward this presents a significant challenge for game developers who must invest additional time and effort to ensure that the models run smoothly without an issue but despite these challenges it is clear that nerve technology has the potential to revolutionize the gaming industry and with the time and continued development it may be possible to create engines that are specifically designed to work with Nerf data Paving the way for a new generation of immersive and realistic gaming experience in fact we have something that is working towards this direction which is the new technology of Unreal Engine 5. that allows us to bring meshes directly to the engine without optimization with topology or UVS which is just amazing so the path ahead might be difficult but the rewards could be extraordinary and it is exciting to think about the possibilities that the future can hold on the other hand when it comes to Nerf generated models it is not on Sunshine and rainbows while these models can look incredibly realistic though are significant challenges that we must overcome first one of the most pressing limitations is the need for a massive amount of training data Jitter with high quality models unfortunately obtaining this gun ethically can be difficult and using stolen data from other artists and people in general is not acceptable at today's standards and a lot of people can find an issue with it additionally some inconsistencies in light in the shadows can be a bit of a headache for users but hey thus a minor issue compared to the downright ugly artifacts that can arise from photogrammetry work and while there have been some exciting Innovations with nervous ability to capture motion scenes there is still a long way to go in this area and some heavy development is still required to make matter worse there aren't a lot of user-friendly software or even platforms for creating nerve generated models but just keep in mind guys that we didn't present these limitations to dishearten any of you it is just a showcase that technology is still brand new and some work is still needed before it can become as efficient as possible nevertheless it is still an exciting and promising Innovation and we are looking forward to witness its Evolution towards becoming the standard and what people actually use it's hard to deny that Nerf technology is a game changer when it comes to 3D while some might argue that it is a replacement I would say it is more of an upgrade think about it North provides all the benefits of photogrammetry with many added benefits and of course it's not like that photogrammetry is gonna disappear overnight since it is a well-established part of many pipelines in different Industries and this is for a good reason since it provides geometry based message directly and it is easy to work with it having said that while traditional photogrammetry may still have a place in some workflows in my opinion 3D scanning may be going in a different direction which is the north Direction and probably in the coming years especially we are seeing the rise of AI there is a great potential for the technology and this may make nerve a dominant technology in 3D scanning and creating 3D assets using data capture technology I hope you guys found this video useful and informative if you did please give it a thumbs up you can also check some of our previous videos thank you very much for watching and I will see you in the next one foreign

Channel: InspirationTuts

Views: 68,891

Keywords: Is Nerf The End of Photogrammetry, nerf, photogrammetry, blender 3d, blender, 2d, 3d animation, 3d modeling, 3d models, gaming, visual effects, architecture

Id: 1eJhsTUEPEg

Length: 11min 16sec (676 seconds)

Published: Fri Jul 21 2023