Paul McComas

869 posts

Paul McComas

Paul McComas

@sucktime

Canuck, drinker of beer, sitter of chairs, 70% water. VP, Head of Animation @ EA

가입일 Mart 2010
395 팔로잉384 팔로워
Paul McComas
Paul McComas@sucktime·
@anorangeduck Mobu should do better exporting bvh…but also bvh is a terrible format that I keep as far away from as possible
English
0
0
1
142
Daniel Holden
Daniel Holden@anorangeduck·
MotionBuilder aggressively rounds floats when it exports BVH files. Unfortunately it doesn't keep enough precision to losslessly encode euler angles close to gimbal. Here is the result of exporting and then re-importing as BVH. You can see the kind of shaking artefacts it makes.
English
3
3
41
6.2K
Paul McComas
Paul McComas@sucktime·
@h4kr Those darn privacy policies for Canada and Europe getting in the way of our corporate overlords I guess
English
1
0
1
42
Paul McComas
Paul McComas@sucktime·
@h4kr On web or something else? Like is Webm any good?
English
1
0
0
70
Scott Graham
Scott Graham@h4kr·
Admittedly, the entirety of everything I know about video codecs I learned in the last two hours, but wow! what a mess. \
English
2
0
0
299
Paul McComas
Paul McComas@sucktime·
I tried the Final Fantasy XVI demo and while so much of it is top notch (visuals, grandeur, etc…) when one element is so much worse than others it really pulls you out of the experience. Did they really look at core loco and say, “yup that’s good enough”? youtu.be/p9O9TmerReU
YouTube video
YouTube
English
0
0
3
315
Paul McComas 리트윗함
AK
AK@_akhaliq·
Example-based Motion Synthesis via Generative Motion Matching paper page: huggingface.co/papers/2306.00… present GenMM, a generative model that "mines" as many diverse motions as possible from a single or few example sequences. In stark contrast to existing data-driven methods, which typically require long offline training time, are prone to visual artifacts, and tend to fail on large and complex skeletons, GenMM inherits the training-free nature and the superior quality of the well-known Motion Matching method. GenMM can synthesize a high-quality motion within a fraction of a second, even with highly complex and large skeletal structures. At the heart of our generative framework lies the generative motion matching module, which utilizes the bidirectional visual similarity as a generative cost function to motion matching, and operates in a multi-stage framework to progressively refine a random guess using exemplar motion matches. In addition to diverse motion generation, we show the versatility of our generative framework by extending it to a number of scenarios that are not possible with motion matching alone, including motion completion, key frame-guided generation, infinite looping, and motion reassembly.
English
6
34
162
93.7K
Paul McComas
Paul McComas@sucktime·
Waiting for my flight to GDC and over hearing a conversation where the phrase “it’s just not part of my brand” was used multiple times. My eyes can’t roll back into my head any further…. 🙄
English
0
0
6
530
Paul McComas
Paul McComas@sucktime·
@anorangeduck It's like my velocity blend! Nice. Cool trick with the slerp shortest arc glitches fixup though.
English
0
0
1
557
Paul McComas
Paul McComas@sucktime·
The character creation tool in Wild Hearts is amazing. So much control and detail. Really well done.
English
0
0
1
279
Paul McComas
Paul McComas@sucktime·
I’ve been enjoying replaying GTA IV recently on the steam deck, an amazing and wonderful experience… however the offline mode is *horrible*. Rockstar is making it genuinely worse than it needs to be. But basically offline mode for GTA doesn’t work and it makes me very sad.
English
0
0
0
261