Friday, 4 May 2012
Evaulation: 04/05/12
Looking at my final piece, there are some aspects that I am pleased with and there are some that are not good enough. At the beginning of the project I knew I would struggle with it as 3D is not my area of interest and I don't have a very good knowledge in it, especially Maya (slightly more knowledge of 3D Studio Max). However when it was explained that the main chunk of the project would be using 123D Catch and MatchMover I thought it might be abit easier, as the programmes both looked fairly simple to grasp and use. When I first tried to create my model in 123D Catch it was pretty much a failure. This was a good thing looking back as it taught me how to take better photographs for use in the programme. I retook my photos and tried the model again. Although it wasn't perfect and needed a lot of cleaning up, I was quite pleased with it. Creating this model successfully pulled my confidence up a little and I felt better baout the project. I then quickly shot my footage and put it into MatchMover.
Here is where the problems began. None of the members in the group released how difficult it is to get a good track form MatchMover. It appeared at first that you just put the footage in the programme and run an automatic track and that was it. But this is not the case and we all began having to do manual tracks to get a decent track. At this point I had a lot of help from Brad who had looked at some online tutorials about MatchMover. He showed me how to put in manual tracking points and where abouts on the footage would be good to place tracks. I was pleased to be learning something new and found putting the points in quite interesting, albeit a little tiresome. I was however a little frustrated at this point because I was suffereing with tonsilitus and didn't have the energy or patience for things to go wrong.
Once we had a decent track I put it into Maya with the model. This caused another set of problems as I struggled to get the footage into Maya, but this was quickly fixed with the slight change of an extension name. When the model and footage was put together in Maya I added a shadow and altered the lighting to make the model look like it was really in the scene. However there was a slight movement, but I was overall pleased with how it looked, as I didn't really think I would be able to reach this stage successfully.
The next stage was to render out of Maya and put all of the elements together in After Effects. For some reason the movement of the model seemed more exagerated in After Effects. By this point I was rather disheartened as everything seemed to have gone wrong with every piece of software we had used. I decided that I would had this in, it was either hand in a piece with a slight wobble or not hand anything in!
Overall I am disappointed that my track isn't good enough and isn't doing the job it should, and if I was to do the project again I would spend more time familiarising myself with MatchMover and creating a better track. However as it goes, I am pleased with how my model turned out and how it looks in the scene.
Here is where the problems began. None of the members in the group released how difficult it is to get a good track form MatchMover. It appeared at first that you just put the footage in the programme and run an automatic track and that was it. But this is not the case and we all began having to do manual tracks to get a decent track. At this point I had a lot of help from Brad who had looked at some online tutorials about MatchMover. He showed me how to put in manual tracking points and where abouts on the footage would be good to place tracks. I was pleased to be learning something new and found putting the points in quite interesting, albeit a little tiresome. I was however a little frustrated at this point because I was suffereing with tonsilitus and didn't have the energy or patience for things to go wrong.
Once we had a decent track I put it into Maya with the model. This caused another set of problems as I struggled to get the footage into Maya, but this was quickly fixed with the slight change of an extension name. When the model and footage was put together in Maya I added a shadow and altered the lighting to make the model look like it was really in the scene. However there was a slight movement, but I was overall pleased with how it looked, as I didn't really think I would be able to reach this stage successfully.
The next stage was to render out of Maya and put all of the elements together in After Effects. For some reason the movement of the model seemed more exagerated in After Effects. By this point I was rather disheartened as everything seemed to have gone wrong with every piece of software we had used. I decided that I would had this in, it was either hand in a piece with a slight wobble or not hand anything in!
Overall I am disappointed that my track isn't good enough and isn't doing the job it should, and if I was to do the project again I would spend more time familiarising myself with MatchMover and creating a better track. However as it goes, I am pleased with how my model turned out and how it looks in the scene.
Deadline day: 04/05/12
Today was the final day for us to work on our projects. I returned to my 3D project with the intentions of improving it. However I have to admit that my knowledge of Maya is limited and I didn't know how to improve the movement of my model. If I had more time I could have gone back to MatchMover and tried to do a more accurate track, but because of illness throughout the week I have not had all of the time I would have liked to work on the project and ask my tutor for help.
When Brad and myself did the manual track of the footage earlier in the week it looked quite good, there were a good number of points and the majority of the track was green. We therefore couldn't understand why the model moved around so much in Maya, when the track we were using appeared to be fairly good.
I decided that I would leave the track as it was, considering the time and would put the model, track and footage together in After Effects. Once I had done this I rendered the video out to hand in.
When Brad and myself did the manual track of the footage earlier in the week it looked quite good, there were a good number of points and the majority of the track was green. We therefore couldn't understand why the model moved around so much in Maya, when the track we were using appeared to be fairly good.
I decided that I would leave the track as it was, considering the time and would put the model, track and footage together in After Effects. Once I had done this I rendered the video out to hand in.
Tuesday, 1 May 2012
Retracking footage and putting model into scene: 01/05/12
Today I continued working on my project. With some help from my course mate Brad we were able to creat a manual track in MatchMover as the automatic track that I did first was nowhere near good enough. It took a long time but we managed to get a decent track of my footage.
I then put the footage and track into Maya and added the barbie model. Once the model was positioned in the place I wanted it I worked with the lighting (which was already quite good) and added a shadow to help it look more realistic.
I then rendered the model and shadow out and put it into After Effects with the original footage. However when I played it back the model was moving around way too much, alot more than it had done in Maya. I asked Brad to look at it but he couldn't figure out why there was a problem exporting the model. I decided to revisit the project tomorrow and try to render out the model again.
I then put the footage and track into Maya and added the barbie model. Once the model was positioned in the place I wanted it I worked with the lighting (which was already quite good) and added a shadow to help it look more realistic.
I then rendered the model and shadow out and put it into After Effects with the original footage. However when I played it back the model was moving around way too much, alot more than it had done in Maya. I asked Brad to look at it but he couldn't figure out why there was a problem exporting the model. I decided to revisit the project tomorrow and try to render out the model again.
Thursday, 26 April 2012
Project Update: 26/04/12
Today I continued to work on my model, playing with the mesh and getting it tidied up ready to go into the scene. I am pretty pleased with how it looks now and think it is ready for the next stage of the project.
I then tried to put my model into my scene, but I have been having some problems with my footage and although I can see the track in Maya, I cannot see the moving footage, which makes it impossible to know whiich tracking points correspond to which piece of footage. I am therefore unable to add my model to the scene at the moment until I have asked my tutor to look at the problem.
Hopefully the problem will be easy to sort out and I will be able to finish my project next week.
I then tried to put my model into my scene, but I have been having some problems with my footage and although I can see the track in Maya, I cannot see the moving footage, which makes it impossible to know whiich tracking points correspond to which piece of footage. I am therefore unable to add my model to the scene at the moment until I have asked my tutor to look at the problem.
Hopefully the problem will be easy to sort out and I will be able to finish my project next week.
Monday, 23 April 2012
3D model:23/04/12
Today I used my new photographs to make my model. The model was much better the second time round and after some manual tracking with help from the tutor I was able to produce a fairly clear model. Once I had this model I exported it into Maya to tidy up.
Once in Maya I edited the mesh of the model and deleted areas of the model that I didn't need, for example the surface the doll was sat on in the photographs.
The next thing I need to do with my model is to put it into my tracked footage and then work on the shadows and lighting to make it look realistic in the scene.
This is going to take a little longer than I first thought as I have decided to reshoot my footage in a different location as the track that I did on the footage of the settee in my living room wasn't very good and the points moved around, which isn't ideal for this project. I am going to shoot new footage of the steps outside of the university this afternoon and track this, as I think this will produce a better track for me to put my model into.
Once in Maya I edited the mesh of the model and deleted areas of the model that I didn't need, for example the surface the doll was sat on in the photographs.
The next thing I need to do with my model is to put it into my tracked footage and then work on the shadows and lighting to make it look realistic in the scene.
This is going to take a little longer than I first thought as I have decided to reshoot my footage in a different location as the track that I did on the footage of the settee in my living room wasn't very good and the points moved around, which isn't ideal for this project. I am going to shoot new footage of the steps outside of the university this afternoon and track this, as I think this will produce a better track for me to put my model into.
Sunday, 22 April 2012
Filming and tracking: 20/04/12
Today I filmed the footage for my project. As we were told to keep the piece simple and my model will be sitting down, I filmed the settee in my conservatory and the settee in my living room. I thought I would do 2 so that I had a choice when it came to putting my model into the scene and could choose the best footage for the best final outcome. I literally panned from one side of the settee to the other.
Later this afternoon I tracked the living room footage using MatchMover read for class on Monday.
I also tried to upload my photographs to 123D Catch to create my model, but the major it of the photographs were too dark and the model wouldn't work. I am going to reshoot the photos this weekend and upload them again in class on Monday.
Later this afternoon I tracked the living room footage using MatchMover read for class on Monday.
I also tried to upload my photographs to 123D Catch to create my model, but the major it of the photographs were too dark and the model wouldn't work. I am going to reshoot the photos this weekend and upload them again in class on Monday.
Monday, 16 April 2012
Project Update: 16/04/12
After my last post my idea changed slightly. I was originally going to model people to put in my scene. However it proved difficult to find a time when my family were available to do the photographs and I was concerned I was running short on time. I therefore decided that I would use dolls for my models. I have photographed 3 barbie dolls which i am going to feed into 123D catch this week to make my models.
After the session with our tutor today I have realised that my idea was way too complex and we are not required to create something as long or difficult as I was thinking. All we have to create is something short, simple but done well. So I am going to shoot the sette in my house and concentrate on making it look as thought the oversized barbie doll model is sat on the settee.
After the session with our tutor today I have realised that my idea was way too complex and we are not required to create something as long or difficult as I was thinking. All we have to create is something short, simple but done well. So I am going to shoot the sette in my house and concentrate on making it look as thought the oversized barbie doll model is sat on the settee.
Example of barbies and bratz dolls used in footage
Thursday, 5 April 2012
3D Text
As well as using 3D models in my project I want to use some 3D text. I have used this in past project and it is really simple to do and looks very effective if done properly. I am planning on using the 3D text to highlight certain words in the song I use, just add a little bit more to the piece.
A couple of examples of 3D text.
References:
http://allbetsareoff.com/tutorials/quick-tip-3d-text/
http://thedesigninspiration.com/fonts/3d-text-render/
http://www.videocopilot.net/tutorials/3d_text_in_3ds_max/
A couple of examples of 3D text.
References:
http://allbetsareoff.com/tutorials/quick-tip-3d-text/
http://thedesigninspiration.com/fonts/3d-text-render/
http://www.videocopilot.net/tutorials/3d_text_in_3ds_max/
Research into Manipulating Live Footage: 05/04/12
Like I mentioned in my previous post, I want to manipulate the footage I shoot for my project to add that extra spark. The videos below demonstrate split screen and slow motion. Unfortunately I couldn't find an example of footage that had been sped up. The videos have chosen are both music videos because I want to add music to my piece. I love music and am often inspired by it, so I want to feature it in my project. n my head at the moment I am considering using some kid of dance track with a fast beat.
Example of Spilt screen
Example of slow motion
Ideas for live footage: 05/04/12
I started thinking about what footage I was going to shoot for my project. I obviously wanted to shoot an environment that I would be able to place my 3D people in. I also knew that I wanted footage that didn't feature any other people, because real people alongside 3D models would make the models look less realistic and I thought this would lower the quality of my project.
Because the piece is only 30 seconds to 1 minute long and the emphasis of the project is on the models, I think that the live footage is not super important so I don't need to spend too long looking for a place to shoot. At the moment I am thinking about my local park, but I will probably go out with my camera and shoot a couple of different locations so I can choose later on.
As well as thinking about the location for the live footage, I also want to do something with the live footage to make it stand out. I want to use the short piece of footage and duplicate it so that it plays on a loop. As well as this I want to play with different speeds, slowing the footage down and then speeding it right up. I am also going to look into split screen. I think the idea of taking one small piece of footage and reusing it in different ways in one piece is an interesting idea and I am going to look into it more.
Because the piece is only 30 seconds to 1 minute long and the emphasis of the project is on the models, I think that the live footage is not super important so I don't need to spend too long looking for a place to shoot. At the moment I am thinking about my local park, but I will probably go out with my camera and shoot a couple of different locations so I can choose later on.
As well as thinking about the location for the live footage, I also want to do something with the live footage to make it stand out. I want to use the short piece of footage and duplicate it so that it plays on a loop. As well as this I want to play with different speeds, slowing the footage down and then speeding it right up. I am also going to look into split screen. I think the idea of taking one small piece of footage and reusing it in different ways in one piece is an interesting idea and I am going to look into it more.
Ideas: 05/04/12
When we had been given the brief, our tutor gave us some advice about things that wouldn't work very well as 3D models. These included trees (because of the leaves), liquid or anything that was transparent of too shiny.
This got me thinking about what I would like to photograph for my 3D model. I wanted to do something that was interesting and that would stand out in the clip. I didn't want to make models of household objects and put them into a house environment, as this seemed pointless. The same applied to the outdoors.
Although I obviously want the 3D models to look like they are part of the scene, I also want them to be the centre of attention and stand out.
I therefore decided quite quickly that I wanted to create 3D people for my scene. I realised this would maybe be more difficult than photographing a still object but after seeing the video in the previous post I thought I could do it. I also want to create more than one model as I think this will add something extra to my piece. However I am aware that my 3D skills aren't brilliant so I may have to stick with using one model in the end, depending on time and how easily I am managing with the project.
I am going to photograph my family, as I would like to have a piece of work that I can keep that has my family in, as they are important to me.
With the decisions made that I will create 3D people and wanting to use live footage in my project, the next step was deciding what footage I will shoot and how I will tie everything together.
This got me thinking about what I would like to photograph for my 3D model. I wanted to do something that was interesting and that would stand out in the clip. I didn't want to make models of household objects and put them into a house environment, as this seemed pointless. The same applied to the outdoors.
Although I obviously want the 3D models to look like they are part of the scene, I also want them to be the centre of attention and stand out.
I therefore decided quite quickly that I wanted to create 3D people for my scene. I realised this would maybe be more difficult than photographing a still object but after seeing the video in the previous post I thought I could do it. I also want to create more than one model as I think this will add something extra to my piece. However I am aware that my 3D skills aren't brilliant so I may have to stick with using one model in the end, depending on time and how easily I am managing with the project.
I am going to photograph my family, as I would like to have a piece of work that I can keep that has my family in, as they are important to me.
With the decisions made that I will create 3D people and wanting to use live footage in my project, the next step was deciding what footage I will shoot and how I will tie everything together.
Researching 123D Catch
After looking at the brief I decided to research 123D Catch as this is a piece of software I have never used before and although I understood the basic concept of the software, I wanted to know more about how it works and how I would need to use it in my project. I visited the 123D Catch website and watched all of the tutorial videos, 2 of which are posted below. I found watching these videos really helpful and realised the software was very simple and that I shouldn't have any problem with it when I come to create my model.
Introducing 123D catch
Generating a model (Buddah example)
After seeing all of the tutorial videos I decided I would search for some examples of work that other people had done. I decided pretty early on that I wanted to make models of people for my project (ideas explained in next blog post), so I wanted to look for videos of people to see if they would work and be as effective as the buddah statue model. I found this video which showed me that if I am careful and thorough when taking my photographs I can produce a good model, like the one below.
Model of a person using 123D Catch
Brief
The brief for this 3D project requires us to use 123D Catch software to create a 3D model or scene to use in a short animation clip. I will create a 3D model as 3D is not one of my strengths and I think a model will be simpler and easier to manage in the project.
We are also required to use other elements in our projects such as 2D animation and graphics, live footage and effects and filters. I will use live footage in my project as production is something I enjoy and editing live footage is my main area of interest. I will also use effects and filters to later my live footage.
The rest of my blog will show the development of my ideas, research and inspirations and the progress of my project.
We are also required to use other elements in our projects such as 2D animation and graphics, live footage and effects and filters. I will use live footage in my project as production is something I enjoy and editing live footage is my main area of interest. I will also use effects and filters to later my live footage.
The rest of my blog will show the development of my ideas, research and inspirations and the progress of my project.
Subscribe to:
Comments (Atom)