- 8
- 149 898
Haotian Zhang
เข้าร่วมเมื่อ 10 ส.ค. 2020
Learning Physically Simulated Tennis Skills from Broadcast Videos
Supplementary video for SIGGRAPH 2023 paper "Learning Physically Simulated Tennis Skills from Broadcast Videos".
For more information, please visit the project website research.nvidia.com/labs/toronto-ai/vid2player3d/
For more information, please visit the project website research.nvidia.com/labs/toronto-ai/vid2player3d/
มุมมอง: 14 128
วีดีโอ
vid2player siggraph long
มุมมอง 1.6K3 ปีที่แล้ว
This is the long presentation video of our SIGGRAPH 2021 (TOG) paper "Vid2Player: Controllable Video Sprites that Behave and Appear like Professional Tennis Players". See the project page at: cs.stanford.edu/~haotianz/research/vid2player/
vid2player siggraph short
มุมมอง 2.3K3 ปีที่แล้ว
This is the short presentation video of our SIGGRAPH 2021 (TOG) paper "Vid2Player: Controllable Video Sprites that Behave and Appear like Professional Tennis Players". See the project page at: cs.stanford.edu/~haotianz/research/vid2player/
Vid2Player: Controllable Video Sprites that Behave and Appear like Professional Tennis Players
มุมมอง 3.2K4 ปีที่แล้ว
This video shows results from the paper "Vid2Player: Controllable Video Sprites that Behave and Appear like Professional Tennis Players". See the project page at: cs.stanford.edu/~haotianz/research/vid2player/ The video you are watching is the latest version with player shadows and shot audio added. You can also find the previous version of the video here: th-cam.com/video/GnZUIuOzgQc/w-d-xo.html
Synthesized fake video between Roger Federer and Novak Djokovic
มุมมอง 9K4 ปีที่แล้ว
This video is synthesized by Vid2Player. For more information, please check our website. cs.stanford.edu/~haotianz/research/vid2player/
Vid2Player: Controllable Video Sprites that Behave and Appear like Professional Tennis Players
มุมมอง 118K4 ปีที่แล้ว
This video shows results from the paper "Vid2Player: Controllable Video Sprites that Behave and Appear like Professional Tennis Players". See the project page at: cs.stanford.edu/~haotianz/research/vid2player/
An Internal Learning Approach to Video Inpainting
มุมมอง 5844 ปีที่แล้ว
Oral presentation for ECCV 2020 workshop on Deep Internal Learning Project website: cs.stanford.edu/~haotianz/research/video_inpainting/
An Internal Learning Approach to Video Inpainting
มุมมอง 5324 ปีที่แล้ว
Supplemental video for ICCV 2019 paper Project website: cs.stanford.edu/~haotianz/research/video_inpainting/
🤯
Hello, this is a great project! Where could I get in touch with you?
Very cool!
Is there a demo?
Amazing!
Mocap meet NOcap
this is good for any racket strategy study and also for others sports such as GOLF
Beautiful work !
It still looks fake. Weird half-volley forehand without full swing.
still fake though. No knockback from when the ball hits the racket. No micro body jerks from movements but only linear speed of movement at the centre of mass. no body tilting to shift centre of mass such as ducking to quickly change dash directions.
Beyond amazing
Great paper and presentation, well done!
even a simulated version of federers strokes is so unique and beautiful
Can we upload this video to our channel is that allowed?
No
@@blenderguy3250 I asked the paper author he said yes :)
@@SECourses but I told you that you can't.
@@blenderguy3250 who ru 😂
good game!
Have you run simulations where your controlled Djokovic sprite is against a low ranked player sprite to see if your simulation is similar to the real match outcome and play style?
Looks like there is now a turntable. Academians are making better games than the game makers :) Use to be gaming industry leading the charge earlier. AI is making the turns!
Good work!
So, soon, I can watch the 202X Super Bowl before it happens!!!!
Aful
Can you actually use this and play it as a tennis game? If so is there a way for us to try it out?
One of the hardest video to edit
The nadal grunt at the beginning 😂
Hi! Do you have a research paper on this ? Would love to read them !
cs.stanford.edu/~haotianz/research/vid2player/
tcwang0509.github.io/vid2vid/paper_vid2vid.pdf
I don't get the point of this video
It's impressive editing
@@Feisty123 gave me a headache
true, the nadal federer at uso I get, but federer and djokovic played multiple times at wimbledon
Future of sports gaming is getting rid of old school animation and instead using AI to create photorealistic stuff which is actually playable. The guy who made this video is way ahead of the game
federer break point simulator
Obviously it's fake. Federer doesn't miss shots like that.
lol yes he does
yeah, usally on match points against
.... Djokovic, mostly in Wimbledon finals
He also doesnt hit deep ones like this and Djoko usually goes near the line but towards the center
Best one yet!
Sick stuff. This is truly incredible work man. Hope it finds its way into a video game
I think you missed the Federer serve with the Nadal Grunt
This needs to become a game
Wow. This is magic
Hi, I am a tennis animator, you should try use your technology in the making of a tennis game in a triple A studio!
Should contact 2K! Steer clear of EA or Big Ben
Bilim teknik... Ekim...2020
BİLİM TEKNİK TEN GELENLER
Ben de ordan geldim
It is like controlling destinity 😅 and I am feeling that reality is losing by day by.
tennis.ai
That`s too much....
Would it be possible to mirror a player? E.g. Nadal becoming right-handed?
1:06 that really hurts Federer fans...
Does it take into account how hard the player hits the bal? (For example Serena hits the ball less hard than Federer. ) Or variation of ball speed within the game?
Who is going to simulate the referee?
Without shadows it looks fake
TO LATE TO THE PARTY . THE GAMBLING HOUSES HAVE ALLREADY MORE ADVANCED SIMULATION THAT THIS.
HOW YOU GUYS DO THIS. WHAT SOFTWARE U USE. IS VOICE ALSO CREATED BY AI
So this one is generated by AI?
Amazing 🤩
want an AR application for this games
Serena is an all-time great female player, but against men she is not in the top 300.
tennis.ai
Looks fake.