As I am using H.264 AVC encoding/decoding technique in my research, so is it good to work? because more advance HEVC/ H.265 is already exist.
I want to know whether H.264 will be still efficient in near future or not?
H.264 is mature technology, a lot of hardware and software exists, including tuned codecs from different companies. H.265 is just started expansion to market. According to recent survey
https://www.encoding.com/files/2016-Global-Media-Formats-Report.pdf
number of files encoded in H.264 files is in 12 times higher than number of files encoded in H.265. So, I suggest not to change codec at least until H.265 takes 25% share.
It depends on the purpose to compress the video i.e upto which extend you want to compress the video. As in all lossy compression, there is a trade-off between video quality, cost of processing the compression and decompression, and system requirements.
Go through these links :
http://www.sciencedirect.com/science/article/pii/S221302091630091X
https://www.researchgate.net/post/Which_is_the_best_techniques_to_compress_video_files
I agree with the two previous posts. These are always difficult questions, during a period of transition. The current trend seems to be that any 4K or 8K video material would use H.265 compression, and I have to believe that when H.265 becomes more commonplace in consumer equipment, H.265 will be used generally. Why not?
But the transition period may be extended. For one thing, just as was the case between H.262 and H.264, older decoders may not be powerful enough to run H.265 decoding, depending what modes are being used. I would expect a long period where material has to be made available both ways.
I got some reasons to work with H.264/AVC, If I am wrong then kindly correct me.
1) H.264 is a well established technique, so it is easy to work with (lots of tools and techniques are available).
2) H.265 does not have sufficient resources (tools) to work with (even don't have much research articles).
3) Most of the good companies are using H.264, and still they will continue use it in near future.
4) I would have to wait until H.265/HEVC does not make a sufficient space in market.
If the goal of the research is to propose new methods or tools to improve compression efficiency of videos, I would recommend you to use the most recent video compression tools when conducting your research and for that H.265 is a good starting point.
But if it is just a black box for you or the focus of the research lies outside the scope of standardisation, the reasons mentioned in your post above may apply.
Just to correct for point 2): There exist sufficient resources to work and there are huge number of research articles on them.
Saurabh Puri, give us a link to free software for conversion of uncompressed avi -> h.265 and vice versa.
I am not aware of any conversion softwares as I work only with the raw (yuv) format for my research but it seems to be do-able using ffmpeg (avi) -> yuv420 -> h.265 and vice versa.
Mr.Sandeep,
As you know that H.264 is a matured technology, it is if you take up research work on H.265/HEVC if you are looking forward to work on video compression.
In my sense, for any publication purpose, you need to compare your video compression result with the current standard ( which is HEVC or H.265). Once I submit a journal (IEEE CSVT) with result comparison with H264 and they directly rejected it as there was no comparison with HEVC.
HEVC latest implementation can be found here https://hevc.hhi.fraunhofer.de/svn/svn_HEVCSoftware/trunk/
and https://hevc.hhi.fraunhofer.de/HM-doc/index.html This one is a good documentation
Hope that helps.
Dear Sandeep Singh Sengar,
to address some of your points:
1. Agreed.
2. HEVC also has lots of tools and literature. For instance, with x265 we are able to achieve real time encoding. Furthermore, there are numerous academic papers and books describing every detail of HEVC.
3. Major companies like Netflix deploy HEVC for their real world applications.
Okay Sir,
May you suggest good tool for H.265, which is easy to understand and in which easy to modify the source code (preferably in MATLAB if not then in C/C++). It will be good if, there is sufficient material to understand the code.
Second question is that:
I have a code of one paper which is implemented in JM reference software (H.264), May I convert it in H.265 using your suggest tool?
For academic research and tool development I suggest HM which is available at https://hevc.hhi.fraunhofer.de/
For other purposes I suggest x265 as modifiable software.
You can also check http://compression.ru/video/codec_comparison/hevc_2015/MSU_HEVC_comparison_2015_free.pdf to learn about some other HEVC implementations.
- Badges
- Science topic
- Similar topics
- Correction