<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=us-ascii"><meta name=Generator content="Microsoft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";
mso-fareast-language:EN-US;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
span.EmailStyle17
{mso-style-type:personal-compose;
font-family:"Calibri","sans-serif";
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri","sans-serif";
mso-fareast-language:EN-US;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]--></head><body lang=EN-CA link=blue vlink=purple><div class=WordSection1><p class=MsoNormal>I’m new to libtheora and video encoding in general, but I have worked hard to educate myself in the basics. I’ve working in image processing for many years, so I’m not starting entirely from scratch. I’m having an encoding problem and I’m looking for helpful suggestions.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>I’m using a very recent build of ffmpeg 0.6 to encode some image frames (+ audio) into theora/vorbis (ogv) streams for HTML5 playback. Not everything supports webm yet and ogv will be a requirement for some time to come.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>My source material starts black, fades up from black over a few seconds and then proceeds. It’s stylized animation, not photo-real photography so backgrounds are fairly flat, contrast is high – basically not the greatest situation to be in for DCT. I’m encoding from image frames, but I’ve also tried transcoding from other video codecs over to theora using ffmpeg/libtheora.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>My problem is this: As the image is fading in during the first shot, the stream hangs for a few moments, then proceeds. It’s noticeable enough to be unacceptable. The moment it hangs, the initial image is perhaps 50% between black and the final (bright) end-point. I notice during the ffmpeg encode that the stream seems to sit at very low bitrates until around 96 frames into the stream, then it jumps up close to the target bitrate I’ve requested, so I believe that it’s an encoding issue and not a player issue. I would have expected the bitrate to ramp up more gradually. <o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Some players don’t even play back the stream, but most demonstrate the same problem. I’m assuming that the “compliant” players all use libtheora and so they are all doing the same thing. I’m using ffmpeg’s player as a baseline because I expect that the encoder and decoder should be somewhat well-matched, though I think that the ffmpeg team might have written their own decoder.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>I suspect it has something to do with the fact that the video fades up from black at the head. The problem does not occur with other input I offer the encoder.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>I’ve tried a variety of settings with ffmpeg, but I don’t believe it exposes all the libtheora options (and I’m not even sure which libtheora options would make a difference). I’ve not been able to find any settings that seem relevant to the problem. I know that h264 has specific handlers to make fade-in and fade-out situations less strenuous on encoders/decoders.<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Does anyone have experience with encoding fade-ins from black at the head of the stream? Is the fade-in really the problem (known issue) or a red herring? <o:p></o:p></p><p class=MsoNormal>Any comments or suggestions?<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p><p class=MsoNormal>Thanks,<o:p></o:p></p><p class=MsoNormal>Mayur<o:p></o:p></p><p class=MsoNormal><o:p> </o:p></p></div></body></html>