« July 2015 | Main | September 2015 »

August 2015

Parametric Humans are Coming to Montreal

Autodesk is sponsoring the third International Workshop on Biomechanical and Parametric Modeling of Human Anatomy (PMHA). As part of the Parametric Human Project, PMHA is all about modeling the human body and simulating it to better understand biomedical and human factors.

Parametric Modeling of Human Anatomy (PMHA)The program for this year's PMHA is rich with content and includes talks on:

  • Developing models for how speech, the tongue and swallowing work
  • Virtual Reality for medical simulation
  • Biomechanical models of the knee and spine
  • Muscle modeling and simulation

This event will take place at Autodesk's beautiful office in Old Montreal.


Autodesk Research at SIGGRAPH 2015

If you're attending SIGGRAPH 2015 in Los Angeles, watch out for these presentations from members of Autodesk Research and say hi to the presenters!:

Kitty: Sketching Dynamic and Interactive Illustrations

Monday, 10 August 3:45 pm - 5:15 pm, Los Angeles Convention Center, Room 402AB

Rubaiat Habib, Fanny Chevalier, Tovi Grossman and George Fitzmaurice

Kitty will be part of the UIST Reprise and you can read more about Kitty in this blog post.

Printing Elastics

Thursday, 13 August 10:45 AM - 12:15 PM, Los Angeles Convention Center, Room 150/151

Session Chair: Nobuyuki Umetani

Convolutional Wasserstein Distances: Efficient Optimal Transportation on Geometric Domains

Tuesday, 11 August 10:45 AM - 12:15 PM, Los Angeles Convention Center, Room 152

Justin Solomon, Fernando de Goes, Gabriel Peyré, Marco Cuturi, Adrian Butscher, Andy Nguyen, Tao Du, Leonidas Guibas

This paper introduces a new class of algorithms for optimization problems involving optimal transportation over geometric domains. The main contribution is to show that optimal transportation can be made tractable over large graphics domains, such as images and triangle meshes, improving performance by orders of magnitude compared to previous work.

OmniAD: Data-Driven Omni-Directional Aerodynamics

Wednesday, 12 August 2:00 PM - 3:30 PM, Los Angeles Convention Center, Room 153A-C

Tobias Martin, Nobuyuki Umetani, Bernd Bickel

Introducing OmniAD, a novel data-driven pipeline to model and acquire the aerodynamics of three-dimensional rigid objects simply by capturing their falling motion using a single camera. OmniAD enables realistic real-time simulation of rigid bodies and interactively designed three-dimensional kites that actually fly.

Computational Tools for 3D Printing

Thursday, 13 August 2:00 pm - 5:15 pm, Los Angeles Convention Center, Room 403AB

Nobuyuki Umetani, Bernd Bickel, Wojciech Matusik

This course reviews current 3D printing hardware and software pipelines, and analyzes their potential and shortcomings. Then it focuses on computational specification for fabrication methods, which allow designing or computing an object's shape and material composition from a functional description.

3D-Printed Prosthetics for the Developing World

Thursday, 13 August, 9-10:30 am,SIGGRAPH Studio

Ryan Schmidt, Ginger Coons, Vincent Chen, Timotheus Gmeiner, Matt Ratto

You can read more about this project here.

Autodesk Exhibitor Sessions

Members of your favourite Autodesk product teams will be showing off cool, new things throughout the week.

The Art of Fluid Animation

And you may see Jos Stam wearing a shirt like this - ask him about it :)

Autodesk Research The Art of Fluid Animation


Five Myths of 3D User Interface Design

Are the things you believe about user interfaces untrue? Maybe something seems logical but you've not really tested it out? Or you've just followed the crowd? Following up on our post about magic and user perception, we are now going to expose the five myths of user interface design.

5 Myths of User Interface Design

Dr. Gordon Kurtenbach, Head of Autodesk Research, has been studying user interfaces in 3D computer graphics for more than two decades and gave a great talk on the myths of user interface design - things he once believed were true but didn't deliver as expected. The five myths of 3D user interface design according to Dr. Kurtenbach are:

  1. 3D Input Devices are Best for Working in 3D
  2. 3D Displays are Best for Working in 3D
  3. Haptic Devices are the Future
  4. Icons = Good Design
  5. A Good User Interface is Natural

Myth #1: 3D Input Devices are Best for Working in 3D

The logic goes that we live in the physical world are always working in 3D. Everyday activities like washing the dishes, folding laundry and driving a car all happen in 3D. The challenge is that when dealing with a computer, there's a piece of glass between the data and the user, resulting in a number of problems:

  • Fatigue: people are used to resting their arms on a desk when dealing with a traditional mouse and keyboard.
  • Positioning is not the only task: 3D input devices are generally used for positioning objects but there is more to working with data than positioning it. For example, there are a lot of data entry type tasks for defining the properties of an object.
  • Depth perception is required: with a piece of glass in the way, we have to find ways to replicate depth. See myth #2 for more info.

Myth #2: 3D Displays are Best for Working in 3D

There's been a lot of work in stereography and many have experienced it in the constrained environment of a movie theatre. Most stereo solutions are faking 3D depth to fool your system. This is what can lead to motion sickness. This is not to say stereo is not well done or worthy of more work - it's just that there are still hurdles to overcome.

Human depth perception is complex. People with one eye can still perceive depth as we rely on a number of cues including the height of the viewer, the height of the viewed objects and the distance from viewer to objects.

The challenges for 3D displays include:

  • Intrusiveness: Viewers must wear glasses or head gear. Some people who don't wear eye glasses do not like to wear glasses. Some who wear glasses have trouble incorporating the additional gear.
  • The quality of display: Both the quality of content and the resolution can lower the viewer's experience. There are also range of view limitations that we don't have in the real world. We can look wherever we want.
  • Tangible benefit: A typical problem for marketers in all industries is explaining how much better the experience gets. Consumers are faced with choices like would you rather a bigger monitor or a smaller stereo monitor?

5 User Interface Myths Autodesk Research

Myth #3: Haptic Devices are the Future

Haptic devices reproduce the sense of touch. Try a quick do-it-yourself haptic device by putting a pen on your computer monitor or something else close by. Trace the contours. How does it feel? Now try the same thing with your finger, the palm of your hand and the back of your hand. How does that feel? It's a richer experience with your finger and hand, isn't it?

Haptic devices are currently only giving basic feedback where our sense of touch is rich. We can get feedback on texture, hardness, temperature, weight, volume, contours, and the shape of the object. Like the myths above, there is a lot of information that we need to replicate in the digital world to make it a meaningful user experience.

One place you can use haptic feedback today is with rapid prototyping. If you were designing a headset, you could 3D print it at scale and try it on.

User Interface Myths Autodesk Research

Myth #4: Icons = Good Design

3D users are visual people, right? And visual people prefer icons. Maybe. But you can get carried away with icons. It's very important not to confuse visual appeal with ease of use.

It's also important not to be lazy and copy the faults of others. Just because it's industry practice to use lots of icons does not mean that lots of icons are good design. Icons are a foreign language and we use pop up tool tips as the translator. To complicate things further, we still rely on antiquated technology to represent some operations. We still use a floppy disk to represent saving. We've got plenty of people in the world who have never seen a floppy disk, let alone used one.

User interface design innovators should seek to improve what exists by taking advantage of the latest technologies. How could one improve upon this situation by using the power of cloud computing? What if tool tips became more visual and played a learning video instead of a line of text? At Autodesk Research, we call it ToolClips!

Autodesk Research ToolClips
ToolClips in AutoCAD provide access to extended documentation and video tutorials

Myth #5: A Good User Interface is Natural

Natural is a tricky word and can be misunderstood. Are we talking about grass and flowers in a meadow? Perhaps a health food store?

We can look at natural as a statement of skills - what do people already have? What experience from the physical world applies to operating a computer? What skills can be transferred from using a web browser to using a word processor? The pillars of direct manipulation provide additional insight:

  • Objects and results should be visible
  • Pointing and moving are strong metaphors
  • Incremental: allow users to work through the process
  • Reversible: allow users to back out of an error
  • Rapid: engage users with an interactive interface

Whatever we call it, we are really trying to accelerate the rate at which novices begin to perform like experts.

Novice to Expert User Interface Transition Autodesk Research
Helping novice users transition to experts

Is something natural the best way to turn novices into experts? A hammer and nail are relatively natural - we've been using tools to hit things for years - but nowadays many experts use a nail gun. It may not be natural but it can sure increase the rate at which someone works.

Marking menus, pictured below, are a great example of the novice (a) to expert (b) transition in software user interfaces. Looking at the expert workflow on the right, the pattern to buy fruit and vegetables looks more like Egyptian Hieroglyphics than a natural expression of grocery shopping in the real world but it is highly efficient. 

Marking Menus Autodesk Research

Summary: Effective user interface technologies are not emulations of the real world

Being able to separate perception from reality is one of the most important things one needs to be able to do when looking at and working with technology. This doesn't mean that a piece of technology only gets one chance. Over time, problems may be resolved and the technology may live up to it's promise. Dr. Kurtenbach expects that he will be proven wrong at some point on the above myths.

Dr. Kurtenbach encourages user interface designers and researchers to invent an exciting future while remembering:

  • Human skills and the ability to learn are powerful and deep
  • Place the human at the centre of design
  • To be a keen observer of the details and the trade-offs
  • Ask and discover what has changed

If you liked this post, please share it with your friends and colleagues using the buttons below.