Hi all,
for the past two months, I have been working on a prototype for a radical
redesign of the swscale internals, specifically the format handling layer.
This includes, or will eventually expand to include, all format input/output
and unscaled special conversion steps.
I am not yet at a point
On date Saturday 2025-03-08 15:30:28 +, Soft Works wrote:
>
>
> > -Original Message-
> > From: Stefano Sabatini
> > Sent: Samstag, 8. März 2025 15:37
> > To: FFmpeg development discussions and patches
> > Cc: Soft Works ; softworkz
> > ; Andreas Rheinhardt
> >
> > Subject: Re: [FFm
..and individual numbering. The benefits are:
* Smaller log file sizes
* The disambiguation is much easier to recognize and to follow
* It eventually allows comparing and viewing log file diffs without almost
every line being different due to those addresses
Before
==
[hevc @ 018e
From: softworkz
which is meant to control prefix formatting. The actual formatting
has to be performed by the consuming application which needs to provide
a custom logging callback via av_log_set_callback().
Signed-off-by: softworkz
---
doc/APIchanges | 3 +++
libavutil/log.h | 5
Peter Ross:
> On Sat, Mar 08, 2025 at 04:29:06AM +0100, Andreas Rheinhardt wrote:
>> Patches attached
>>
>> - Andreas
>
>
> no objections, lgtm
>
> -- Peter
> (A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
Thanks. Will apply these patches tonight unless there are objections.
- Andreas
__
Patch attached.
- Andreas
From 6a826c7b15021eeb8c4b3ab288262c18daa43fcf Mon Sep 17 00:00:00 2001
From: Andreas Rheinhardt
Date: Sun, 9 Mar 2025 04:52:50 +0100
Subject: [PATCH] avcodec/snow: Remove ff_snow_release_buffer()
Pointless after 7e41f95dce6390f39a5134a25213828ed65fac6b.
Signed-off-by:
Zhao Zhili:
> From: Zhao Zhili
>
> BSF can update extradata, e.g., vvc_mp4toannexb. If we don't copy
> bsf->par_out back to avcodec context, decoder can get extradata in
> mp4 format, while packets are in annexb format.
> ---
> libavcodec/decode.c | 3 +++
> 1 file changed, 3 insertions(+)
>
>
> On Mar 9, 2025, at 11:06, Andreas Rheinhardt
> wrote:
>
> Zhao Zhili:
>> From: Zhao Zhili
>>
>> BSF can update extradata, e.g., vvc_mp4toannexb. If we don't copy
>> bsf->par_out back to avcodec context, decoder can get extradata in
>> mp4 format, while packets are in annexb format.
>> ---
Andreas Rheinhardt:
> Patches attached
>
> - Andreas
Will apply the two sbcenc patches tomorrow unless there are objections.
- Andreas
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscri
Hi,
On Sat, Mar 8, 2025 at 9:58 AM wrote:
>
> Those functions will be used by classify in the upcoming patches.
>
> Try the new filters using my Github Repo
> https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
>
> Any Feedback is appreciated!
>
> Signed-off-by: MaximilianKaindl
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 15:46
> To: FFmpeg development discussions and patches
> Cc: Soft Works ; softworkz
> ; Andreas Rheinhardt
>
> Subject: Re: [FFmpeg-devel] [PATCH v3 3/7] fftools/ffprobe: Rename
> writer_print_section_* and Wr
From: softworkz
Signed-off-by: softworkz
---
fftools/Makefile | 12 +
fftools/ffprobe.c | 1849 -
2 files changed, 142 insertions(+), 1719 deletions(-)
diff --git a/fftools/Makefile b/fftools/Makefile
index 4499799818..664b73b161 100644
--- a/ffto
From: softworkz
Signed-off-by: softworkz
---
fftools/textformat/avtextformat.c | 672 +
fftools/textformat/avtextformat.h | 171
fftools/textformat/avtextwriters.h | 68 +++
fftools/textformat/tf_compact.c| 282
fftools/textformat/tf_defa
From: softworkz
separated for better clarity of the preceding commit
Signed-off-by: softworkz
ren
---
fftools/ffprobe.c | 361 +++---
1 file changed, 178 insertions(+), 183 deletions(-)
diff --git a/fftools/ffprobe.c b/fftools/ffprobe.c
index f398057df7
This new version of the patchset starts by extracting the text formatting
and writing APIs from ffprobe.c into a subfolder under fftools. The type
naming follows public API naming style, ramping up for making it a public
API in the future without another big renaming.
The extraction of the text fo
From: softworkz
Signed-off-by: softworkz
---
fftools/ffprobe.c | 534 +++---
1 file changed, 267 insertions(+), 267 deletions(-)
diff --git a/fftools/ffprobe.c b/fftools/ffprobe.c
index 4a90bc4824..406bef4d2f 100644
--- a/fftools/ffprobe.c
+++ b/fftools/
From: softworkz
Enables filtergraph printing and adds the options to the docs
Signed-off-by: softworkz
---
doc/ffmpeg.texi | 10 ++
fftools/ffmpeg.c| 4
fftools/ffmpeg_filter.c | 5 +
3 files changed, 19 insertions(+)
diff --git a/doc/ffmpeg.texi b/doc/ffmpe
From: softworkz
to allow print_graph to access the information.
Signed-off-by: softworkz
---
fftools/ffmpeg_filter.c | 188 +---
fftools/ffmpeg_filter.h | 232
2 files changed, 233 insertions(+), 187 deletions(-)
create mode
From: softworkz
---
doc/APIchanges | 3 +++
libavfilter/avfilter.c | 9 +
libavfilter/avfilter.h | 12
libavfilter/version.h | 2 +-
4 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/doc/APIchanges b/doc/APIchanges
index 5a64836e25..354716399d 100644
From: softworkz
The key benefits are:
- Different to other graph printing methods, this is outputting:
- all graphs with runtime state
(including auto-inserted filters)
- each graph with its inputs and outputs
- all filters with their in- and output pads
- all connections between all
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 15:46
> To: FFmpeg development discussions and patches
> Cc: Soft Works ; softworkz
> ; Andreas Rheinhardt
>
> Subject: Re: [FFmpeg-devel] [PATCH v3 3/7] fftools/ffprobe: Rename
> writer_print_section_* and Wri
From: softworkz
This commit adds the memaddresses log flag.
When specifying this flag at the command line, context prefixes will
be printed with memory addresses like in earlier ffmpeg versions.
Example with memaddresses flag:
[hevc @ 018e72a89cc0] .
without (new behavior):
[hevc #0]
From: softworkz
This goes together with a change to logging of context prefixes, which
is printing logical ids instead of memory addresses.
The benefits are:
- Smaller log file sizes
- The disambiguation is much easier to recognize and to follow
- It eventually allows comparing and viewing log f
From: softworkz
---
doc/fftools-common-opts.texi | 2 ++
1 file changed, 2 insertions(+)
diff --git a/doc/fftools-common-opts.texi b/doc/fftools-common-opts.texi
index f6d452c40e..756c843c02 100644
--- a/doc/fftools-common-opts.texi
+++ b/doc/fftools-common-opts.texi
@@ -230,6 +230,8 @@ log to
Signed-off-by: James Almer
---
libavfilter/drawutils.c | 2 ++
libavfilter/vsrc_testsrc.c| 19 +++
tests/fate/filter-video.mak | 6 ++
tests/ref/fate/filter-yuvtestsrc-nv24 | 10 ++
tests/ref/fate/filter-yuvtestsrc-p410 | 10 ++
On 6 Mar 2025, at 18:44, Soft Works wrote:
>> -Original Message-
>> From: ffmpeg-devel On Behalf Of Marvin
>> Scholz
>> Sent: Donnerstag, 6. März 2025 18:38
>> To: FFmpeg development discussions and patches
>> Subject: Re: [FFmpeg-devel] [PATCH] avutil/log: Replace addresses in log
>>
Andreas Rheinhardt:
> Andreas Rheinhardt:
>> Patches attached
>>
>
> And here is another one on top of this that gets rid of all the
> deprecation warnings with Clang.
>
> - Andreas
>
Will apply this patch tomorrow unless there are objections.
- Andreas
___
> -Original Message-
> From: ffmpeg-devel On Behalf Of Soft
> Works
> Sent: Freitag, 7. März 2025 05:36
> To: FFmpeg development discussions and patches
> Subject: [FFmpeg-devel] The Concept for the CC Installment is broken by
> Design
[..]
> The Community Committee Concept is broken
Performance wise, encoding a 1080p 1-minute video is performed in about 2.5
minutes with the cpu encoder running on my Ryzen 5 4600H, while it takes about
30 seconds on my NVIDIA GTX 1650
Haar shader has a subgroup optimized variant that applies when configured
wavelet depth allows it
lavapipe
---
libavcodec/vulkan/vc2_dwt_haar.comp | 70 +++
libavcodec/vulkan/vc2_dwt_haar_subgroup.comp | 89 +
libavcodec/vulkan/vc2_dwt_hor_legall.comp| 66 +++
libavcodec/vulkan/vc2_dwt_upload.comp| 29 +++
libavcodec/vulkan/vc2_dwt_ver_legall.comp| 62 +
---
libavcodec/vulkan/common.comp | 58 +++
1 file changed, 46 insertions(+), 12 deletions(-)
diff --git a/libavcodec/vulkan/common.comp b/libavcodec/vulkan/common.comp
index e4e983b3e2..3dc1527529 100644
--- a/libavcodec/vulkan/common.comp
+++ b/libavcodec/vulkan/
From: IndecisiveTurtle
---
libavcodec/Makefile| 2 +-
libavcodec/vc2enc.c| 515 +
libavcodec/vc2enc_common.c | 321 +++
libavcodec/vc2enc_common.h | 323 +++
4 files changed, 653 insertions(+), 508 dele
On date Saturday 2025-03-01 10:01:58 +, softworkz wrote:
[...]
> +int avtext_context_open(AVTextFormatContext **ptctx, const AVTextFormatter
> *formatter, AVTextWriterContext *writer, const char *args,
> +const struct AVTextFormatSection *sections, int
> nb_sections,
On date Saturday 2025-03-01 10:02:00 +, softworkz wrote:
> From: softworkz
>
> separated for better clarity of the preceding commit
>
> Signed-off-by: softworkz
> ren
> ---
> fftools/ffprobe.c | 361 +++---
> 1 file changed, 178 insertions(+), 183 de
Those functions will be used by classify in the upcoming patches.
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
libavutil/detection_bbox.c | 54 ++
Implements batch tokenization support using the tokenizers-cpp library,
providing functions to load tokenizers and encode text batches. This is crucial
for CLIP/CLAP models that need to process text prompts.
https://github.com/mlc-ai/tokenizers-cpp
Try the new filters using my Github Repo
htt
Defines new DNNFunctionType enums for CLIP and CLAP inference and adds new data
structures like DNNExecZeroShotClassificationParams to support zero-shot
classification models.
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedb
Moves the contain_valid_detection_bbox function from the OpenVINO backend to
the common backend code, making it available for all DNN backends to use when
checking bounding box validity.
Will be used by the Torch backend in an upcoming patch in this series.
Try the new filters using my Github R
Extends the DNN filter common code to support CLIP/CLAP classification and
model loading with tokenizers. Adds new execution functions for both image and
audio classification.
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedba
This major patch completely rewrites the dnn_classify filter, transforming it
from a video-only filter into a versatile media classification system that
supports:
Standard image classification (via OpenVINO)
CLIP-based image classification (via LibTorch)
CLAP-based audio classification (via LibT
This patch introduces a new avgclass filter that aggregates classification
results across multiple frames or audio segments. Key features:
Works with both video and audio streams
Collects and averages classification probabilities over time
Exports results both to logs and optional CSV files
Suppo
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
doc/filters.texi | 170 +++
1 file changed, 85 insertions(+), 85 deletions(-
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 15:01
> To: FFmpeg development discussions and patches
> Cc: Soft Works ; softworkz
> ; Andreas Rheinhardt
>
> Subject: Re: [FFmpeg-devel] [PATCH v3 1/7] fftools/textformat: Extract
> and generalize textformat
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
doc/filters.texi | 64
1 file changed, 64 insertions(+)
diff --git a/d
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
libavfilter/dnn/dnn_backend_torch.cpp | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/libavfilte
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
configure | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/configure b/configure
index b2a19488
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
configure | 2 ++
1 file changed, 2 insertions(+)
diff --git a/configure b/configure
index 2c08901adc..b2a1948855 10075
On date Wednesday 2025-02-19 20:07:55 +, softworkz wrote:
> From: softworkz
>
> Signed-off-by: softworkz
> ---
> fftools/ffprobe.c | 147 +-
> 1 file changed, 146 insertions(+), 1 deletion(-)
Should this be postponed to the ffprobe.c formats refa
IndecisiveTurtle:
> From: IndecisiveTurtle
>
> ---
> libavcodec/Makefile| 2 +-
> libavcodec/vc2enc.c| 515 +
> libavcodec/vc2enc_common.c | 321 +++
> libavcodec/vc2enc_common.h | 323 +++
> 4 files ch
Andreas Rheinhardt:
> Patch attached.
>
> - Andreas
>
Another patch attached
- Andreas
From 37d77b0820c19ea5519769d13a3e54ba88234c0e Mon Sep 17 00:00:00 2001
From: Andreas Rheinhardt
Date: Sat, 8 Mar 2025 17:35:32 +0100
Subject: [PATCH 2/2] avcodec/put_bits: Add and use put_bits63()
When usin
> a) Don't top-post.
Sorry my mistake, I hope this message is proper now
> b) If you put a static array in a header, it will be included in every
> file that actually uses it and therefore end up being duplicated in the
> binary.
Ah I understand now, thanks will also fix that.
Hi Manuel
On Tue, Mar 04, 2025 at 06:07:17PM +0100, Manuel Lauss wrote:
> Don't error out, just ignore unknown codec numbers and pretend
> decode succeeded. This is useful for older LucasArts titles
> which stack a lot of different FOBJs with different codecs into
> a single frame.
>
> Signed-of
On date Saturday 2025-03-08 17:55:29 +, softworkz wrote:
> From: softworkz
>
> Signed-off-by: softworkz
> ---
> fftools/Makefile | 12 +
> fftools/ffprobe.c | 1849 -
> 2 files changed, 142 insertions(+), 1719 deletions(-)
Should still be good
On date Saturday 2025-03-08 17:55:30 +, softworkz wrote:
> From: softworkz
>
> separated for better clarity of the preceding commit
>
> Signed-off-by: softworkz
> ren
> ---
> fftools/ffprobe.c | 361 +++---
> 1 file changed, 178 insertions(+), 183 de
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 19:12
> To: Soft Works
> Cc: FFmpeg development discussions and patches de...@ffmpeg.org>; Soft Works
> Subject: Re: [FFmpeg-devel] [PATCH v3 1/7] fftools/textformat: Extract
> and generalize textformat api fr
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 20:09
> To: FFmpeg development discussions and patches
> Cc: softworkz
> Subject: Re: [FFmpeg-devel] [PATCH v5 1/8] fftools/textformat: Extract
> and generalize textformat api from ffprobe.c
>
> On date Satur
Hi Manuel
On Tue, Mar 04, 2025 at 06:07:18PM +0100, Manuel Lauss wrote:
> The left and top parameters of an FOBJ are signed values.
>
> Signed-off-by: Manuel Lauss
> ---
> v4: revert v3, it arose due to a misunderstanding
> v3: change the bytestream accessor to signed too
> v2: no changes
> lib
On date Saturday 2025-03-08 17:55:28 +, softworkz wrote:
> From: softworkz
>
> Signed-off-by: softworkz
> ---
> fftools/textformat/avtextformat.c | 672 +
> fftools/textformat/avtextformat.h | 171
> fftools/textformat/avtextwriters.h | 68 +++
> ffto
On Fri, Mar 07, 2025 at 01:36:03AM +0100, Michael Niedermayer wrote:
> Avoids out of array access
> Another solution may be choosen later if something else turns out to fit
> naturally into the decoder. But for now dont crash
>
> Signed-off-by: Michael Niedermayer
> ---
> libavcodec/ffv1dec_temp
Hi all
As we effectively have now 3 implementations of RAW LSB support, some feedback
would be welcome!
latest is:
0307 1:36 To FFmpeg devel ( 14K) [FFmpeg-devel] [PATCH 1/2] libavcodec/ffv1:
Support storing decorrelated LSB raw without rangecoder
The first was a hack basically, the 2nd went t
From: softworkz
Signed-off-by: softworkz
---
fftools/Makefile | 12 +
fftools/ffprobe.c | 1849 -
2 files changed, 142 insertions(+), 1719 deletions(-)
diff --git a/fftools/Makefile b/fftools/Makefile
index 4499799818..664b73b161 100644
--- a/ffto
From: softworkz
to allow print_graph to access the information.
Signed-off-by: softworkz
---
fftools/ffmpeg_filter.c | 188 +---
fftools/ffmpeg_filter.h | 232
2 files changed, 233 insertions(+), 187 deletions(-)
create mode
From: softworkz
separated for better clarity of the preceding commit
Signed-off-by: softworkz
ren
---
fftools/ffprobe.c | 361 +++---
1 file changed, 178 insertions(+), 183 deletions(-)
diff --git a/fftools/ffprobe.c b/fftools/ffprobe.c
index f398057df7
From: softworkz
---
doc/APIchanges | 3 +++
libavfilter/avfilter.c | 9 +
libavfilter/avfilter.h | 12
libavfilter/version.h | 2 +-
4 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/doc/APIchanges b/doc/APIchanges
index 5a64836e25..354716399d 100644
From: softworkz
The key benefits are:
- Different to other graph printing methods, this is outputting:
- all graphs with runtime state
(including auto-inserted filters)
- each graph with its inputs and outputs
- all filters with their in- and output pads
- all connections between all
On Thu, Mar 06, 2025 at 12:43:56AM +0100, Michael Niedermayer wrote:
> This improves compression by 0.1% overall and 44% for the changed table
> I tried several other things but so far this is the best
> compromise between complexity and compression
>
> This can also be extended to 32 and 64bit fl
From: softworkz
Enables filtergraph printing and adds the options to the docs
Signed-off-by: softworkz
---
doc/ffmpeg.texi | 10 ++
fftools/ffmpeg.c| 4
fftools/ffmpeg_filter.c | 5 +
3 files changed, 19 insertions(+)
diff --git a/doc/ffmpeg.texi b/doc/ffmpe
On Sat, Mar 08, 2025 at 04:29:06AM +0100, Andreas Rheinhardt wrote:
> Patches attached
>
> - Andreas
> From 54e27b588e18fdfe277a77b0f385cb02731022b9 Mon Sep 17 00:00:00 2001
> From: Andreas Rheinhardt
> Date: Thu, 6 Mar 2025 17:56:08 +0100
> Subject: [PATCH 1/4] avcodec/vp8: Fix wrong #endif com
This new version of the patchset starts by extracting the text formatting
and writing APIs from ffprobe.c into a subfolder under fftools. The type
naming follows public API naming style, ramping up for making it a public
API in the future without another big renaming.
The extraction of the text fo
From: softworkz
Signed-off-by: softworkz
---
fftools/textformat/avtextformat.c | 672 +
fftools/textformat/avtextformat.h | 171
fftools/textformat/avtextwriters.h | 68 +++
fftools/textformat/tf_compact.c| 282
fftools/textformat/tf_defa
From: softworkz
Signed-off-by: softworkz
---
fftools/ffprobe.c | 512 +++---
1 file changed, 256 insertions(+), 256 deletions(-)
diff --git a/fftools/ffprobe.c b/fftools/ffprobe.c
index 4a90bc4824..9f869b41b3 100644
--- a/fftools/ffprobe.c
+++ b/fftools/
This will make it possible to transmit raw video over NUT without losing
the field order information.
---
libavformat/nutdec.c | 4
libavformat/nutenc.c | 3 +++
2 files changed, 7 insertions(+)
diff --git a/libavformat/nutdec.c b/libavformat/nutdec.c
index 34b7e3cb9a..07eff23134 100644
---
Hello Andreas
Thank you for your feedback. I will resubmit the patch with the proper
implementation you described.
Could you elaborate on the array duplication comment? I don't see any other
place it's defined
Στις Σάβ 8 Μαρ 2025 στις 3:14 μ.μ., ο/η Andreas Rheinhardt <
andreas.rheinha...@outlook
Sorry for delayed review, due to a sickness on my side for the past
three days.
On date Saturday 2025-03-01 10:01:58 +, softworkz wrote:
> From: softworkz
>
> Signed-off-by: softworkz
> ---
> fftools/textformat/avtextformat.c | 671 +
> fftools/textformat/avtex
IndecisiveTurtle:
> Hello Andreas
>
> Thank you for your feedback. I will resubmit the patch with the proper
> implementation you described.
> Could you elaborate on the array duplication comment? I don't see any other
> place it's defined
>
> Στις Σάβ 8 Μαρ 2025 στις 3:14 μ.μ., ο/η Andreas Rhein
On date Saturday 2025-03-01 10:01:59 +, softworkz wrote:
> From: softworkz
>
> Signed-off-by: softworkz
> ---
> fftools/Makefile | 12 +
> fftools/ffprobe.c | 1849 -
> 2 files changed, 142 insertions(+), 1719 deletions(-)
[...]
Overall makes
> On 2 Mar 2025, at 20:38, Soft Works
> wrote:
>
>
>
>> -Original Message-
>> From: ffmpeg-devel On Behalf Of Ronan
>> Waide
>> Sent: Sonntag, 2. März 2025 18:24
>> To: ffmpeg-devel@ffmpeg.org
>> Cc: Ronan Waide
>> Subject: [FFmpeg-devel] [PATCH v5] libavcodec/dvbsubenc.c: add a
>>
> -Original Message-
> From: ffmpeg-devel On Behalf Of Marvin
> Scholz
> Sent: Donnerstag, 6. März 2025 19:59
> To: FFmpeg development discussions and patches
> Subject: Re: [FFmpeg-devel] [PATCH] avutil/log: Replace addresses in log
> output with simple ids
>
>
>
> On 6 Mar 2025, at
This substantial patch implements the LibTorch backend support for CLIP
(Contrastive Language-Image Pre-training) and CLAP (Contrastive Language-Audio
Pre-training) models. Key features include:
- Text tokenization and processing for language-based classification
- Support for both image and aud
On date Wednesday 2025-02-19 20:07:56 +, softworkz wrote:
> From: softworkz
>
> Signed-off-by: softworkz
> ---
> doc/ffprobe.texi | 16
> 1 file changed, 16 insertions(+)
>
> diff --git a/doc/ffprobe.texi b/doc/ffprobe.texi
> index ceedc8793b..106d865e18 100644
> --- a/doc
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
libavfilter/avf_dnn_classify.c| 2 +-
libavfilter/dnn/dnn_backend_torch.cpp | 66 ---
2
Try the new filters using my Github Repo
https://github.com/MaximilianKaindl/DeepFFMPEGVideoClassification.
Any Feedback is appreciated!
Signed-off-by: MaximilianKaindl
---
doc/filters.texi | 106 +--
1 file changed, 76 insertions(+), 30 deletions(-
Patch attached.
- Andreas
From 4e50b2adb6de16754b629eaf18fc92d67ecdf1ca Mon Sep 17 00:00:00 2001
From: Andreas Rheinhardt
Date: Sat, 8 Mar 2025 15:45:24 +0100
Subject: [PATCH] avcodec/vc2enc: Simplify writing dirac golomb codes
The earlier code used a loop to determine the number of bits used
an
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 15:37
> To: FFmpeg development discussions and patches
> Cc: Soft Works ; softworkz
> ; Andreas Rheinhardt
>
> Subject: Re: [FFmpeg-devel] [PATCH v3 1/7] fftools/textformat: Extract
> and generalize textformat
Performance wise, encoding a 1080p 1-minute video is performed in about 2.5
minutes with the cpu encoder running on my Ryzen 5 4600H, while it takes about
30 seconds on my NVIDIA GTX 1650
Haar shader has a subgroup optimized variant that applies when configured
wavelet depth allows it
lavapipe
From: IndecisiveTurtle
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 499f826635..a96c700745 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -768,7 +768,7 @@ OBJS-$(CONFIG_VC1_CUVID_DECODER) += cuviddec.o
OBJS-$(CONFIG_VC1_MMAL_DECODER)+= mmaldec.o
OBJS
> -Original Message-
> From: Stefano Sabatini
> Sent: Samstag, 8. März 2025 16:06
> To: FFmpeg development discussions and patches
> Cc: softworkz
> Subject: Re: [FFmpeg-devel] [PATCH 2/2] docs: Add documentation about
> YAML output writer
>
> On date Wednesday 2025-02-19 20:07:56 +0
---
libavcodec/vulkan/vc2_dwt_haar.comp | 66 +++
libavcodec/vulkan/vc2_dwt_haar_subgroup.comp | 89 +
libavcodec/vulkan/vc2_dwt_hor_legall.comp| 66 +++
libavcodec/vulkan/vc2_dwt_upload.comp| 29 +++
libavcodec/vulkan/vc2_dwt_ver_legall.comp| 62 +
88 matches
Mail list logo
