From: "J Bills" Received: from [40.92.7.52] (HELO NAM03-CO1-obe.outbound.protection.outlook.com) by media-motion.tv (CommuniGate Pro SMTP 6.1.0) with ESMTPS id 6467453 for AE-List@media-motion.tv; Wed, 20 Jun 2018 09:37:49 +0200 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=hotmail.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=Qn/B1t2n6IHuob2+caTI6NmBvQODFFastLwFrlGRimY=; b=libLl0+JiTkkJrr+Wql2yB+6bhMzlvziUlnorxAX7jRQB2j4LUny78r4AnYwf/yQAJ1ZQAtRyLcfTaTEqT0dn+LMbmzRVI6aXIw0EcBkxdtDYXRfUAVsPaJD7Wncstrm/I72RW1N08ZsZPcTzLS6RLADdCel1Hr6HwbRNwkpQU28iYbpll95GcEYSuatpjNCsXuL9HeDuC6llePwp0KYlcEJWoZDeZ2vs9UsPOXMgU2oG4omeoJCJBm6LUGXf9v88qfz75rd2FI4g0CzocY2zUccwi830u5LT7I3FHEBWVp/5hh1JDKV0N2aWAPsHlbFRCFsYHn6Sa+sS9b6GAOVUA== Received: from BY2NAM03FT028.eop-NAM03.prod.protection.outlook.com (10.152.84.59) by BY2NAM03HT061.eop-NAM03.prod.protection.outlook.com (10.152.85.133) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA384_P384) id 15.20.884.17; Wed, 20 Jun 2018 07:45:29 +0000 Received: from MWHPR11MB1504.namprd11.prod.outlook.com (10.152.84.53) by BY2NAM03FT028.mail.protection.outlook.com (10.152.84.238) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256_P256) id 15.20.884.17 via Frontend Transport; Wed, 20 Jun 2018 07:45:29 +0000 Received: from MWHPR11MB1504.namprd11.prod.outlook.com ([fe80::490:2f24:b4b9:4f14]) by MWHPR11MB1504.namprd11.prod.outlook.com ([fe80::490:2f24:b4b9:4f14%12]) with mapi id 15.20.0863.016; Wed, 20 Jun 2018 07:45:29 +0000 To: After Effects Mail List Subject: Re: [AE] nVidias Slo mo demo Thread-Topic: [AE] nVidias Slo mo demo Thread-Index: AQHUCGFmDw2eBIMLMkqeMEV5c5I3I6RovGH3 Date: Wed, 20 Jun 2018 07:45:29 +0000 Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-incomingtopheadermarker: OriginalChecksum:CD3AB1C84B7A4BC3E6A2D0A70648BC459BD9A2A4A7F3F427269647031B19EE70;UpperCasedChecksum:3B200E02BDCB88E81D81191D72510C01084A02B94376970474B1A43204502124;SizeAsReceived:6945;Count:46 x-ms-exchange-messagesentrepresentingtype: 1 x-tmn: [btW41sUQWz+2iq/Qlid3iV4fYEVe9g2D] x-ms-publictraffictype: Email x-microsoft-exchange-diagnostics: 1;BY2NAM03HT061;7:vzXTxeq+JiQ/bJxdyV0yUdB3q0mVS3Qbp/Sl259TDPEjKLClLQ3Se6w9hW7ofNIejWusOuvk1VBA+uoGAB+mwq6s28I7cvNjxMrTSj439KSJhffHcCB+i2tgTUytgN8Ny2P4YKlK7DO9jdOSM5MJ7l7VZdBbd8KReuGZ1UTBeJTP7qKN7cVpIIktNWj0TRhowx/zRehFx4k63WgtVok7tA15ijg8wZ5OB8Nxc8sMtAT4kFQjW/6R7gWDHfUv+3xW x-incomingheadercount: 46 x-eopattributedmessage: 0 x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:(7020095)(201702061078)(5061506573)(5061507331)(1603103135)(2017031320274)(2017031324274)(2017031323274)(2017031322404)(1603101448)(1601125500)(1701031045);SRVR:BY2NAM03HT061; x-ms-traffictypediagnostic: BY2NAM03HT061: x-exchange-antispam-report-cfa-test: BCL:0;PCL:0;RULEID:(444000031);SRVR:BY2NAM03HT061;BCL:0;PCL:0;RULEID:;SRVR:BY2NAM03HT061; x-forefront-prvs: 070912876F x-forefront-antispam-report: SFV:NSPM;SFS:(7070007)(199004)(189003)(86362001)(76176011)(99286004)(476003)(7696005)(486006)(14454004)(102836004)(19627405001)(74316002)(53546011)(446003)(11346002)(6506007)(59450400001)(966005)(6346003)(83332001)(81156014)(5250100002)(105586002)(8676002)(3280700002)(68736007)(8936002)(3660700001)(106356001)(6606003)(5660300001)(229853002)(606006)(2900100001)(104016004)(6916009)(33656002)(25786009)(87572001)(26005)(82202002)(97736004)(73972006)(6436002)(55016002)(20460500001)(236005)(54896002)(6306002)(9686003)(6246003)(15852004);DIR:OUT;SFP:1901;SCL:1;SRVR:BY2NAM03HT061;H:MWHPR11MB1504.namprd11.prod.outlook.com;FPR:;SPF:None;PTR:InfoNoRecords;A:1;MX:1;LANG:; received-spf: None (protection.outlook.com: hotmail.com does not designate permitted sender hosts) authentication-results: spf=none (sender IP is ) smtp.mailfrom=djbills@hotmail.com; x-microsoft-antispam-message-info: OfAVKPdHy0GkUPr1nbCQAC/6NVrE7VeG8pXNFO6ZfvMQlcPCOUjx+f3WOSZPmQVRA2iAcIylb5G4KCiixGwWUfZiYnvZ0q7IcEfECN3Yl6Tu2Tjd9bpnRURTOGfD9fInzY0KAtHLFwjEmnoNSxXgQ849jb9zIZDMyDH8+7QS/jXbVBa+atfuTIn+QyZwgq8j Content-Type: multipart/alternative; boundary="_000_MWHPR11MB1504D1974B5F81AF6B0CF4E8B9770MWHPR11MB1504namp_" MIME-Version: 1.0 X-OriginatorOrg: hotmail.com X-MS-Exchange-CrossTenant-RMS-PersistedConsumerOrg: 24fd1209-d934-423e-a578-ee886993c07f X-MS-Exchange-CrossTenant-Network-Message-Id: 06f3f531-5aeb-4444-be74-08d5d681cb75 X-MS-Exchange-CrossTenant-rms-persistedconsumerorg: 24fd1209-d934-423e-a578-ee886993c07f X-MS-Exchange-CrossTenant-originalarrivaltime: 20 Jun 2018 07:45:29.3998 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Internet X-MS-Exchange-CrossTenant-id: 84df9e7f-e9f6-40af-b435-aaaaaaaaaaaa X-MS-Exchange-Transport-CrossTenantHeadersStamped: BY2NAM03HT061 --_000_MWHPR11MB1504D1974B5F81AF6B0CF4E8B9770MWHPR11MB1504namp_ Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Totally agree with you - seems to be nothing special, on par with current s= olutions. Sometimes works, sometimes has tons of artifacts. I respect the= m for baring it and not just showing ideal footage. Twixtor, Kronos, etc are all pretty old in the tooth now to be honest. Man= y studios have their own in-house optical flow systems that produce much mu= ch better results. Denser vector data sets. Better algorithms and edge ha= ndling, far less "fraying" for lack of a better word. Ghosting. This tech really needs a revisit and more "pro" offerings since many of us = have enough computer horsepower at our disposal to go for the better result= s. And it's not just for retiming - there are lots of other tools and use = cases that can use optical flow for tracking/paint/roto or to accelerate ce= rtain processes. There is a lot of damage to be done here, and Foundry doi= ng the Smart Vector tools over in Nuke are really just the tip of the icebe= rg. They could be smarter. har har. GPU acceleration is great, and you definitely notice when you're on a machi= ne that doesn't have a video card up to par to use the GPU acceleration on,= say Kronos for example. So it should definitely be on Nvidia's radar. B= ut whoever goes for the gold and produces the eventual system that will pro= duce the level of results I'm after will understand that this is something = that has to be distributed to a *real* CPU render farm at 1 frame per node,= and cooked at length. Long render times, comparing many many more frames = than the current systems are using. Want! ________________________________ From: After Effects Mail List on behalf of Chris = Zwar Sent: Tuesday, June 19, 2018 10:38:29 PM To: After Effects Mail List Subject: [AE] nVidias Slo mo demo Yesterday nVidia posted a demo of their slo-mo research, which uses AI / ma= chine learning to dramatically slow down footage and interpolate between fr= ames. https://www.youtube.com/watch?time_continue=3D43&v=3DMjViy6kyiqs I have to say that I don=92t really see any advantage over existing optical= flow technology (ie twixtor). I did make the brave step of actually readi= ng the YouTube comments, and a few people did mention twixtor with one pers= on saying the nVidia version is =93vastly superior=94. But I don=92t see i= t. Similar artefacts, lots of blurring, I just don=92t see an immediate ad= vantage over existing technology. Of course it=92s difficult to judge with= out access to the source footage itself. Twixtor and comparable plugins ca= n vary widely depending on the quality of the source footage, with things l= ike noise/grain and compression artefacts sometimes being a make or break f= actor. So maybe this new technology is more tolerant? What do you think? -Chris --_000_MWHPR11MB1504D1974B5F81AF6B0CF4E8B9770MWHPR11MB1504namp_ Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable

Totally agree with you - seems to= be nothing special, on par with current solutions.  Sometimes works, = sometimes has tons of artifacts.  I respect them for baring it an= d not just showing ideal footage.


Twixtor, Kronos, etc are all pret= ty old in the tooth now to be honest.  Many studios have their own in-= house optical flow systems that produce much much better results.  Den= ser vector data sets.  Better algorithms and edge handling, far less "fraying" for lack of a better word.&nbs= p; Ghosting.


This tech really needs a revisit = and more "pro" offerings since many of us have enough computer ho= rsepower at our disposal to go for the better results.  And it's not j= ust for retiming - there are lots of other tools and use cases that can use optical flow for tracking/paint/roto or to acce= lerate certain processes.  There is a lot of damage to be done here, a= nd Foundry doing the Smart Vector tools over in Nuke are really just the ti= p of the iceberg.  They could be smarter.  har har.


GPU acceleration is great, and yo= u definitely notice when you're on a machine that doesn't have a video card= up to par to use the GPU acceleration on, say Kronos for example.  So= it should definitely be on Nvidia's radar.   But whoever goes for the gold and produces the eventual system that= will produce the level of results I'm after will understand that this is s= omething that has to be distributed to a *real* CPU render farm at 1 frame = per node, and cooked at length.  Long render times, comparing many many more frames than the current systems are= using.  Want!



From: After Effects Mail Li= st <AE-List@media-motion.tv> on behalf of Chris Zwar <AE-List@medi= a-motion.tv>
Sent: Tuesday, June 19, 2018 10:38:29 PM
To: After Effects Mail List
Subject: [AE] nVidias Slo mo demo
 

https://www.youtube.com/watch?time_continu= e=3D43&v=3DMjViy6kyiqs

I have to say that I don=92t really see any advantage over = existing optical flow technology (ie twixtor).  I did make the brave s= tep of actually reading the YouTube comments, and a few people did mention = twixtor with one person saying the nVidia version is =93vastly superior=94.  But I don=92t see it.  Simila= r artefacts, lots of blurring, I just don=92t see an immediate advantage ov= er existing technology.  Of course it=92s difficult to judge without a= ccess to the source footage itself.  Twixtor and comparable plugins can vary widely depending on the quality of the source footage, wi= th things like noise/grain and compression artefacts sometimes being a make= or break factor.  So maybe this new technology is more tolerant?

What do you think?

-Chris



--_000_MWHPR11MB1504D1974B5F81AF6B0CF4E8B9770MWHPR11MB1504namp_--