- 14
- 77 941
Jungdam Won
Приєднався 8 бер 2012
[SIGGRAPH 2024] Strategy and Skill Learning for Physics-based Table Tennis Animation
jiashunwang.github.io/PhysicsPingPong/
Переглядів: 259
Відео
[SIGGRAPH 2022] Physics-based Character Controllers Using Conditional VAEs
Переглядів 8 тис.2 роки тому
[SIGGRAPH 2022] Physics-based Character Controllers Using Conditional VAEs High-quality motion capture datasets are now publicly available, and researchers have used them to create kinematics-based controllers that can generate plausible and diverse human motions without conditioning on specific goals (i.e., a task-agnostic generative model). In this paper, we present an algorithm to build such...
Control Strategies for Physically Simulated Characters Performing Two-player Competitive Sports
Переглядів 2,5 тис.3 роки тому
ProjectCode: FairPlay Publication Jungdam Won, Deepak Gopinath, and Jessica Hodgins, Control Strategies for Physically Simulated Characters Performing Two-player Competitive Sports, ACM Transactions on Graphics (SIGGRAPH 2021). Abstract In two-player competitive sports, such as boxing and fencing, athletes often demonstrate efficient and tactical movements during a competition. In this paper, w...
A Scalable Approach to Control Diverse Behaviors for Physically Simulated Characters
Переглядів 3,7 тис.4 роки тому
Project Page research. publications/a-scalable-approach-to-control-diverse-behaviors-for-physically-simulated-characters/ Abstract Human characters with a broad range of natural looking and physically realistic behaviors will enable the construction of compelling interactive experiences. In this paper, we develop a technique for learning controllers for a large set of heterogeneous behavi...
Learning Body Shape Variation in Physics-based Characters (SIGGRAPH Asia 2019)
Переглядів 3,1 тис.5 років тому
Project Page mrl.snu.ac.kr/publications/ProjectMorphCon/MorphCon.html Abstract Recently, deep reinforcement learning (DRL) has attracted great attention in designing controllers for physics-based characters. Despite the recent success of DRL, the learned controller is viable for a single character. Changes in body size and proportions require learning controllers from scratch. In this paper, we...
Aerobatics Control of Flying Creatures via Self-Regulated Learning (SIGGRAPH ASIA 2018)
Переглядів 5506 років тому
Project Page mrl.snu.ac.kr/research/ProjectAerobatics/Aerobatics.htm Abstract Flying creatures in animated films often perform highly dynamic aerobatic maneuvers, which require their extreme of exercise capacity and skillful control. Designing physics-based controllers (a.k.a., control policies) for aerobatic maneuvers is very challenging because dynamic states remain in unstable equilibrium mo...
Aerobatics Control of Flying Creatures via Self-Regulated Learning (SIGGRAPH ASIA 2018)
Переглядів 1,8 тис.6 років тому
Project Page mrl.snu.ac.kr/research/ProjectAerobatics/Aerobatics.htm Abstract Flying creatures in animated films often perform highly dynamic aerobatic maneuvers, which require their extreme of exercise capacity and skillful control. Designing physics-based controllers (a.k.a., control policies) for aerobatic maneuvers is very challenging because dynamic states remain in unstable equilibrium mo...
How to Train Your Dragon: Example-Guided Control of Flapping Flight
Переглядів 7 тис.7 років тому
Project Page mrl.snu.ac.kr/research/ProjectDragon/Dragon.htm Abstract Imaginary winged creatures in computer animation applications are expected to perform a variety of motor skills in a physically realistic and controllable manner. Designing physics-based controllers for a flying creature is still very challenging particularly when the dynamic model of the creatures is high-dimensional, having...
Shadow Theatre: Discovering Human Motion from a Sequence of Silhouettes - Supplemental
Переглядів 1,5 тис.8 років тому
mrl.snu.ac.kr/research/ProjectShadowTheatre/ShadowTheatre.htm Abstract Shadow theatre is a genre of performance art in which the actors are only visible as shadows projected on the screen. The goal of this study is to generate animated characters, the shadows of which match a sequence of target silhouettes. This poses several challenges. The motion of multiple characters are carefully coordinat...
Shadow Theatre: Discovering Human Motion from a Sequence of Silhouettes (SIGGRAPH 2016)
Переглядів 43 тис.8 років тому
mrl.snu.ac.kr/research/ProjectShadowTheatre/ShadowTheatre.htm Abstract Shadow theatre is a genre of performance art in which the actors are only visible as shadows projected on the screen. The goal of this study is to generate animated characters, the shadows of which match a sequence of target silhouettes. This poses several challenges. The motion of multiple characters are carefully coordinat...
Generating and Ranking Diverse Multi-Character Interactions
Переглядів 2,7 тис.10 років тому
mrl.snu.ac.kr/research/ProjectPrevis/preVis.htm Abstract In many application areas, such as animation for pre-visualizing movie sequences or choreography of dancing or other show performances, only a high-level description of the desired scene is provided as input, either written or verbal. Such sparsity, however, lends itself well to the creative process, as the choreographer, animator or dire...
Generating and Ranking Diverse Multi-Character Interactions
Переглядів 82310 років тому
mrl.snu.ac.kr/research/ProjectPrevis/preVis.htm Abstract In many application areas, such as animation for pre-visualizing movie sequences or choreography of dancing or other show performances, only a high-level description of the desired scene is provided as input, either written or verbal. Such sparsity, however, lends itself well to the creative process, as the choreographer, animator or dire...
Data-Driven Control of Flapping Flight
Переглядів 1,9 тис.11 років тому
mrl.snu.ac.kr/research/ProjectBird/birdFlight.htm Abstract We present a physically based controller that simulates the flapping behavior of a bird in flight.We recorded the motion of a dove using marker-based optical motion capture and high-speed video cameras. The bird flight data thus acquired allow us to parameterize natural wingbeat cycles and provide the simulated bird with reference traje...
Data-Driven Control of Flapping Flight Supplemental
Переглядів 67311 років тому
mrl.snu.ac.kr/research/ProjectBird/birdFlight.htm Abstract We present a physically based controller that simulates the flapping behavior of a bird in flight.We recorded the motion of a dove using marker-based optical motion capture and high-speed video cameras. The bird flight data thus acquired allow us to parameterize natural wingbeat cycles and provide the simulated bird with reference traje...
0:05
0:13
What engine are you using?
Anyway to get this working in unreal engine?
Is it possible to make two actors fighting?
I also would like to know if there is some good way to do physics based interaction between characters. The common approach is not physics based, is just playing animation on each character and match their position
Great work!
Amazing! is this configurable with different Mass of characters with existing models?
I guess existing models might work for slight mass change to some extent similarly to other physics-based controllers. However, to get the best performance, the controller needs to be trained for the new character.
WOW
I like your channel. You have a new subscriber! Do you know of promosm?!? You should use it to grow your channel!
Hey man, is this open for the public? Am i able to test ai on this environment?
Where can we access The Interactive Character Creator ? :)
Anyone else here from two minute papers?
yup
I'm on day three of a research binge inspired by that OpenAI Hide and Seek video that Two Minute Papers covered.
So exciting research! May I try the software or get the project sources? Thanks very much
우왕 갓 정담...
Amazing! Do you have any plan of release of the program or code?
Awesome! I like this project.
Wow very impressive work! May I ask what software was used for the physics simulation/animation? Thank you :)
pierpaolo lucarelli Thank you for watching! DART was used for the simulation, UNREAL was used for the rendering. You can find more detail in our paper / project webpage.
Great, Thank you so much for answering :)
This is the best vid about this topic that ive ever seen.
Asshole simulator 2017
When/how does the system decide to use props to complete a target shape?
Zergling Probably based on the presence of too-thin features (wheel spokes, knobs on the violin neck) or other cases where acceptable shapes cannot be found (plane wings).