<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic H263 rate control skipping almost all the frames in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900708#M12704</link>
    <description>&lt;P&gt;Hi there,&lt;/P&gt;
&lt;P&gt;I am using H263 codec with Ipp5.3 gold to encode a YUV file 182.yuv which I will provide as an attachment. I am using a bitrate of 50000 bps. I have frame skippling enabled.&lt;/P&gt;
&lt;P&gt;There are 168 frames in the file, after encode almost 150 frames are skipped, pnly 18 are remaining. This is totally unacceptable Am i doing something wrong here. The YUV file is not too complex in motion, it is a news clip.&lt;/P&gt;
&lt;P&gt;I am unable to attach the file as it is 25 MB. Is there a frp server i can upload to.&lt;/P&gt;
&lt;P&gt;My encoder parameters are as follows :&lt;/P&gt;&lt;FONT size="2"&gt;
&lt;P&gt;30000 // frame scale&lt;/P&gt;
&lt;P&gt;1001 // frame interval&lt;/P&gt;
&lt;P&gt;7 // IVOP quant&lt;/P&gt;
&lt;P&gt;7 // PVOP quant&lt;/P&gt;
&lt;P&gt;9 // BVOP quant&lt;/P&gt;
&lt;P&gt;10 // IVOP distance&lt;/P&gt;
&lt;P&gt;1 // PVOP distance&lt;/P&gt;
&lt;P&gt;15 // PVOP searchWidth in full pel&lt;/P&gt;
&lt;P&gt;15 // PVOP searchHeight in full pel&lt;/P&gt;
&lt;P&gt;1 // Motion Estimation algorithm: 0 - full search, 1 - fast search&lt;/P&gt;
&lt;P&gt;2 // Motion Estimation accuracy: 1 - full pel, 2 - half pel&lt;/P&gt;
&lt;P&gt;0// calculate PSNR&lt;/P&gt;
&lt;P&gt;1// RateControl 0 - no, 1 - FrameLayer(VBR)&lt;/P&gt;
&lt;P&gt;50000 // BitRate, Kbps&lt;/P&gt;
&lt;P&gt;50 // SceneChangeThreshold, %&lt;/P&gt;
&lt;P&gt;1// unrestricted motion vector: 0 - off; 1 - no PLUSPTYPE; 2 - PLUSPTYPE, UUI==1; 3 - PLUSPTYPE, UUI==01&lt;/P&gt;
&lt;P&gt;0// advanced prediction: 0 - off (1MV), 1 - on (4MV, OBMC)&lt;/P&gt;
&lt;P&gt;0// advanced Intra coding -- Annex I support &lt;/P&gt;
&lt;P&gt;0// insert GOB headers: 0 - off, 1 - GOB headers without stuffing, 2 - GOB headers with stuffing (GSTUF)&lt;/P&gt;
&lt;P&gt;0// modified quantization mode: 0 - off, 1 - on -- Annex T&lt;/P&gt;
&lt;P&gt;1 // frame skipping: 0 - disabled, 1 - enabled, default 1&lt;/P&gt;&lt;/FONT&gt;
&lt;P&gt;Please help,&lt;/P&gt;
&lt;P&gt;thanks,&lt;/P&gt;
&lt;P&gt;Ram&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
    <pubDate>Tue, 22 Apr 2008 15:49:22 GMT</pubDate>
    <dc:creator>ramachandran_ramani</dc:creator>
    <dc:date>2008-04-22T15:49:22Z</dc:date>
    <item>
      <title>H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900708#M12704</link>
      <description>&lt;P&gt;Hi there,&lt;/P&gt;
&lt;P&gt;I am using H263 codec with Ipp5.3 gold to encode a YUV file 182.yuv which I will provide as an attachment. I am using a bitrate of 50000 bps. I have frame skippling enabled.&lt;/P&gt;
&lt;P&gt;There are 168 frames in the file, after encode almost 150 frames are skipped, pnly 18 are remaining. This is totally unacceptable Am i doing something wrong here. The YUV file is not too complex in motion, it is a news clip.&lt;/P&gt;
&lt;P&gt;I am unable to attach the file as it is 25 MB. Is there a frp server i can upload to.&lt;/P&gt;
&lt;P&gt;My encoder parameters are as follows :&lt;/P&gt;&lt;FONT size="2"&gt;
&lt;P&gt;30000 // frame scale&lt;/P&gt;
&lt;P&gt;1001 // frame interval&lt;/P&gt;
&lt;P&gt;7 // IVOP quant&lt;/P&gt;
&lt;P&gt;7 // PVOP quant&lt;/P&gt;
&lt;P&gt;9 // BVOP quant&lt;/P&gt;
&lt;P&gt;10 // IVOP distance&lt;/P&gt;
&lt;P&gt;1 // PVOP distance&lt;/P&gt;
&lt;P&gt;15 // PVOP searchWidth in full pel&lt;/P&gt;
&lt;P&gt;15 // PVOP searchHeight in full pel&lt;/P&gt;
&lt;P&gt;1 // Motion Estimation algorithm: 0 - full search, 1 - fast search&lt;/P&gt;
&lt;P&gt;2 // Motion Estimation accuracy: 1 - full pel, 2 - half pel&lt;/P&gt;
&lt;P&gt;0// calculate PSNR&lt;/P&gt;
&lt;P&gt;1// RateControl 0 - no, 1 - FrameLayer(VBR)&lt;/P&gt;
&lt;P&gt;50000 // BitRate, Kbps&lt;/P&gt;
&lt;P&gt;50 // SceneChangeThreshold, %&lt;/P&gt;
&lt;P&gt;1// unrestricted motion vector: 0 - off; 1 - no PLUSPTYPE; 2 - PLUSPTYPE, UUI==1; 3 - PLUSPTYPE, UUI==01&lt;/P&gt;
&lt;P&gt;0// advanced prediction: 0 - off (1MV), 1 - on (4MV, OBMC)&lt;/P&gt;
&lt;P&gt;0// advanced Intra coding -- Annex I support &lt;/P&gt;
&lt;P&gt;0// insert GOB headers: 0 - off, 1 - GOB headers without stuffing, 2 - GOB headers with stuffing (GSTUF)&lt;/P&gt;
&lt;P&gt;0// modified quantization mode: 0 - off, 1 - on -- Annex T&lt;/P&gt;
&lt;P&gt;1 // frame skipping: 0 - disabled, 1 - enabled, default 1&lt;/P&gt;&lt;/FONT&gt;
&lt;P&gt;Please help,&lt;/P&gt;
&lt;P&gt;thanks,&lt;/P&gt;
&lt;P&gt;Ram&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
      <pubDate>Tue, 22 Apr 2008 15:49:22 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900708#M12704</guid>
      <dc:creator>ramachandran_ramani</dc:creator>
      <dc:date>2008-04-22T15:49:22Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900709#M12705</link>
      <description>&lt;P&gt;Hi Ram,&lt;/P&gt;
&lt;P&gt;What is the frame size of the stream?&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Timofei&lt;/P&gt;</description>
      <pubDate>Wed, 23 Apr 2008 09:19:24 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900709#M12705</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2008-04-23T09:19:24Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900710#M12706</link>
      <description>&lt;P&gt;timofei,&lt;/P&gt;
&lt;P&gt;The YUV input isof size CIF. The max frame size generate is about 4189 bytes, but almost all frames are skipped.&lt;/P&gt;
&lt;P&gt;does that answer your question ? Do you think i can upload the YUV file somewhere.&lt;/P&gt;
&lt;P&gt;Ram&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
      <pubDate>Wed, 23 Apr 2008 17:15:45 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900710#M12706</guid>
      <dc:creator>ramachandran_ramani</dc:creator>
      <dc:date>2008-04-23T17:15:45Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900711#M12707</link>
      <description>&lt;P&gt;Ram, can you please try to upload the stream to &lt;/P&gt;
&lt;P&gt;&lt;A&gt;ftp://ftp.intel.com/pub/incoming/&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;If, for any reasonm this doesn't work, you can encode the stream with high quality settings into any format and attach it. I believe the skip-frame issue won't be gone after stream compression/decompression.&lt;/P&gt;</description>
      <pubDate>Fri, 25 Apr 2008 08:47:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900711#M12707</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2008-04-25T08:47:21Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900712#M12708</link>
      <description>&lt;P&gt;OK.. I have uploaded the file. The filename is 182.yuv&lt;/P&gt;
&lt;P&gt;It is in YUV 420 format. I am using H263 encoder, I have specified the encoder options in the previous post.&lt;/P&gt;
&lt;P&gt;Also considering that there are many networks wired and wireless, supporting different packet payload types, there should be provision for limiting the encoded frame sizes.&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Ram&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
      <pubDate>Fri, 25 Apr 2008 15:15:27 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900712#M12708</guid>
      <dc:creator>ramachandran_ramani</dc:creator>
      <dc:date>2008-04-25T15:15:27Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900713#M12709</link>
      <description>&lt;P&gt;Got it. Will investigate. &lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Timofei&lt;/P&gt;</description>
      <pubDate>Fri, 25 Apr 2008 16:22:00 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900713#M12709</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2008-04-25T16:22:00Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900714#M12710</link>
      <description>&lt;P&gt;Hi Ram,&lt;/P&gt;
&lt;P&gt;With the parameters you use (I-frame distance set to 10 being the key one) encoding the stream into 50000 bps is impossible: encoding an all-black (DC only) Intra MB requires 53 bits, which for 352x388 frame makes 20988 bits, thus for 3 Intra frames per second (30/10) the resultant bitrate will inevitably exceed 63000 bps. Setting the I-frame distance to a value &amp;gt; 168 reduces the number of skipped frames considerably: 66 frames get encoded. With the rate control (and frame skipping) off and quantizers set to 31 (maximum), the stream is encoded with the bitrate exceeding the target 50000 bps by a factor of ~2.5. Which means that the stream can not be encoded into 50000 bps without dropping a considerable number of frames, unless some extra-low-bitrate tricks like forced MB skipping or transform coefficients zeroing are employed (and they are not impemented in our encoder).&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Timofei&lt;/P&gt;</description>
      <pubDate>Mon, 28 Apr 2008 13:17:56 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900714#M12710</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2008-04-28T13:17:56Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900715#M12711</link>
      <description>&lt;P&gt;Timofei,&lt;/P&gt;
&lt;P&gt;These techniques like forced MB skipping or zeroing transform coefficients, are they implemented in the MPEG4 rate control or H264 rate control.&lt;/P&gt;
&lt;P&gt;if so, can it be brought into H263, or it needs a lot of modification.&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Ram&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
      <pubDate>Mon, 28 Apr 2008 15:10:38 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900715#M12711</guid>
      <dc:creator>ramachandran_ramani</dc:creator>
      <dc:date>2008-04-28T15:10:38Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900716#M12712</link>
      <description>&lt;P&gt;Ram, &lt;/P&gt;
&lt;P&gt;No, there are no such things in MPEG4/H264 rate control. Actually, they're not a real panacea - the impact on the quality is very serious.&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Timofei&lt;/P&gt;</description>
      <pubDate>Wed, 30 Apr 2008 10:54:50 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900716#M12712</guid>
      <dc:creator>Intel_C_Intel</dc:creator>
      <dc:date>2008-04-30T10:54:50Z</dc:date>
    </item>
    <item>
      <title>Re: H263 rate control skipping almost all the frames</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900717#M12713</link>
      <description>&lt;P&gt;Timofei,&lt;/P&gt;
&lt;P&gt;For the same example I sent you, using the bitrate as 384K with 30 fps. I do MPEG4 encoding with IVOPdist as 150 i.e everey 5 seconds. I get a frame of size 24568 bytes for the frame immediately after the black frame, although it should have been encoded as an I-frame, it was encoded as a P-frame with huge size. I am using a scene change threshold of 50.&lt;/P&gt;
&lt;P&gt;why is this happening. The same encoding for a bitrate of 50 Kbps encodes the frame immediately after the black frame as an I-frame.&lt;/P&gt;
&lt;P&gt;Please throw some light on resolving this issue.&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Ram &lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;
&lt;P&gt;&lt;/P&gt;</description>
      <pubDate>Wed, 14 May 2008 14:01:05 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/H263-rate-control-skipping-almost-all-the-frames/m-p/900717#M12713</guid>
      <dc:creator>ramachandran_ramani</dc:creator>
      <dc:date>2008-05-14T14:01:05Z</dc:date>
    </item>
  </channel>
</rss>

