[Bf-blender-cvs] [17a96051cf0] master: Re-bundle Libmv to ensure it is in sync with the repo
Sergey Sharybin
noreply at git.blender.org
Thu Oct 21 12:41:00 CEST 2021
Commit: 17a96051cf0f664509638bc31b714a4925b5052c
Author: Sergey Sharybin
Date: Thu Oct 21 12:39:00 2021 +0200
Branches: master
https://developer.blender.org/rB17a96051cf0f664509638bc31b714a4925b5052c
Re-bundle Libmv to ensure it is in sync with the repo
===================================================================
M intern/libmv/CMakeLists.txt
M intern/libmv/ChangeLog
M intern/libmv/bundle.sh
M intern/libmv/files.txt
M intern/libmv/mkfiles.sh
M intern/libmv/third_party/msinttypes/inttypes.h
M intern/libmv/third_party/msinttypes/stdint.h
===================================================================
diff --git a/intern/libmv/CMakeLists.txt b/intern/libmv/CMakeLists.txt
index 91ef1f4d038..596e8efa082 100644
--- a/intern/libmv/CMakeLists.txt
+++ b/intern/libmv/CMakeLists.txt
@@ -156,6 +156,7 @@ if(WITH_LIBMV)
libmv/base/scoped_ptr.h
libmv/base/vector.h
libmv/base/vector_utils.h
+ libmv/build/build_config.h
libmv/image/array_nd.h
libmv/image/convolve.h
libmv/image/correlation.h
@@ -199,6 +200,7 @@ if(WITH_LIBMV)
libmv/simple_pipeline/reconstruction_scale.h
libmv/simple_pipeline/resect.h
libmv/simple_pipeline/tracks.h
+ libmv/threading/threading.h
libmv/tracking/brute_region_tracker.h
libmv/tracking/hybrid_region_tracker.h
libmv/tracking/kalman_filter.h
diff --git a/intern/libmv/ChangeLog b/intern/libmv/ChangeLog
index 81096dd90c9..80e69878c99 100644
--- a/intern/libmv/ChangeLog
+++ b/intern/libmv/ChangeLog
@@ -1,520 +1,556 @@
-commit efd7a93317e0278b99e66785f667823e451daef1
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Tue May 9 10:16:42 2017 +0200
+commit f701b2b9fbc641b3252b3513239eeb14477ed5e1
+Author: Lazydodo <github at lazydodo.com>
+Date: Wed Aug 21 10:38:33 2019 -0600
- Fix strict compiler warnings, unused variables
+ Cleanup: Fix build error with MSVC
+
+ Previously eigens internal include order somehow implicitly provided
+ M_PI and friends. The recent eigen version bump broke this implicit
+ behaviour, better to be explicit that we need the math defines for MSVC.
-commit 8efd47e13dfdd3f7209bc96f26d0b13127dd6376
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed Dec 14 10:44:57 2016 +0100
+commit 5b2892f5a3cb6a7a2300f1c32e93d90e22302286
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Mon Mar 15 14:58:13 2021 +0100
- Fix T50243: libmv_panography_test is broken
-
- There was fully wrong logic in comparison: was actually accessing memory
- past the array boundary. Run test manually and the figure seems correct
- to me now.
-
- Spotted by @LazyDodo, thanks!
+ Cleanup: Spelling in Libmv comments
-commit 6dfb9cd1bd14669d84be789000ce234747fb00ff
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Thu Jul 14 11:49:38 2016 +0200
+commit 54cdd2fe5cad29df83cb4d3c3ef1e02e06b022a7
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Thu Oct 21 12:09:30 2021 +0200
- Fix some strict compiler warnings
-
- One of them was a real bug!
+ Ensure clang-format is disabled for 3rd party libraries
-commit f61adaecf7b29ebe6677be0e1c825f0a8d475e4b
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed May 31 11:22:34 2017 +0200
+commit 517d51654d6a4cf790a340b657cc0f7dc00f3158
+Author: Campbell Barton <ideasman42 at gmail.com>
+Date: Tue Jul 13 21:58:45 2021 +1000
- Enable explicit schur complement for BA step
+ Fix x/y mismatch in retract region tracker
+
+ Correct X/Y mismatch in RetrackRegionTracker.
- This is something we do in Blender and only reason it was not
- enabled for standalone Libmv is because we did not have fresh
- enough version of Ceres bundled.
+ NOTE: This isn't used at the moment.
+
+ Reviewed By: sergey
+
+ Ref D11895
-commit fc5d3a1d4880c6658aff693c1c1e8c10c96ce1a7
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed Nov 2 15:32:11 2016 +0100
+commit cfcfc803cf599c3bd7bbfa3bac7d5e9c8bca284b
+Author: Jesse Yurkovich <jesse.y at gmail.com>
+Date: Mon Jul 12 21:01:18 2021 -0700
- Update tests to make tests pass after recent Ceres update
+ Cleanup: Use correct _WIN32/64 defines for MSVC
- Just a precision issue, difference is around 1e-7. Should be fine to
- simply update expected value.
+ Docs: https://docs.microsoft.com/en-us/cpp/preprocessor/predefined-macros
+
+ Differential Revision: https://developer.blender.org/D11460
-commit e1ac9f6124110c1a90d8e417bea47acfcbdcca42
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed May 31 10:54:48 2017 +0200
+commit ad8dfd41958b0e4cb90c7c4e5cc20f93c0832e73
+Author: Campbell Barton <ideasman42 at gmail.com>
+Date: Thu Jun 24 15:56:58 2021 +1000
- Update Ceres to latest release 1.12.0
+ Cleanup: comment blocks, trailing space in comments
-commit ac1571352b4962f110929b963f8616d7310ceea5
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Fri Apr 7 17:10:44 2017 +0200
+commit 9a6cc5daa6c38f33ce2385fe489c07d97586d452
+Author: Campbell Barton <ideasman42 at gmail.com>
+Date: Sun Jun 13 15:11:40 2021 +1000
- Fix crash of keyframe selection on 32bit linux
+ Fix new[]/delete[] mismatch
-commit 5f8df3da965686df39a6ae5c9f17482075017bf4
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Tue Jan 19 14:00:53 2016 +0500
+commit 5591d12928df74ed3517b5e61eeda0b64e6ade4f
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Mon Mar 22 15:16:07 2021 +0100
- Solve some strict warnings in tests
+ Fix T86591: Tracking backwards is slower
+
+ The root of the issue was caused by the PredictMarkerPosition()
+ always returning false when tracking backwards. This was making
+ it so tracker always had to run brute initialization, which is
+ an expensive operation.
+
+ From own timing here:
+
+ - Tracking forward takes 0.667637 seconds
+ - Tracking backward used to take 2.591856 seconds
+ - Tracking backward now takes 0.827724 seconds
+
+ This is a very nice speedup, although the tracking backwards is
+ still somewhat slower. Will be investigated further as part of
+ a regular development.
-commit 8ea3a5d752a9ce3337ab7643897472a4d33747f1
-Author: Brecht Van Lommel <brechtvanlommel at gmail.com>
-Date: Sat Feb 18 23:52:31 2017 +0100
+commit ad9546d2319b0db3e1ccc41c4f84899729d1ad1e
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Mon Mar 15 15:48:15 2021 +0100
- Fix a few compiler warnings with macOS / clang.
+ Fix T86262: Tracking backwards fails after gap in track
+
+ The issue was caused by a prediction algorithm detecting tracking the
+ wrong way. Solved by passing tracking direction explicitly, so that
+ prediction will always happen correctly regardless of the state of the
+ Tracks context.
-commit ffbe81461770e70736e80b8cab8e6eb1f8b27160
-Author: Mike Erwin <significant.bit at gmail.com>
-Date: Wed May 31 10:43:08 2017 +0200
+commit 0773a8d6dfe1bf997b8e42d61d136a65b67fde88
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Thu Oct 21 11:18:10 2021 +0200
- Fix comparison of identicals
+ Cleanup: clang-format
- Some of these check that dimensions match before running code that
- assumes they do match.
+ Is based on Google style which was used in the Libmv project before,
+ but is now consistently applied for the sources of the library itself
+ and to C-API. With some time C-API will likely be removed, and it
+ makes it easier to make it follow Libmv style, hence the diversion
+ from Blender's style.
- Found with PVS-Studio T48917.
-
-commit 206c01999cde16c1c6c43a8e13ffa86020821d98
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed May 31 10:39:16 2017 +0200
-
- Add basic track masking API in place
+ There are quite some exceptions (clang-format off) in the code around
+ Eigen matrix initialization. It is rather annoying, and there could be
+ some neat way to make initialization readable without such exception.
- This brings back ability to mask non-interesting parts of
- specific track (the feature got lost with new auto-track API).
+ Could be some places where loss of readability in matrix initialization
+ got lost as the change is quite big. If this has happened it is easier
+ to address readability once actually working on the code.
- Added it back by extending frame accessor class. This isn't really
- a frame thing, but we don't have other type of accessor here.
+ This change allowed to spot some missing header guards, so that's nice.
- Surely, we can use old-style API here and pass mask via region
- tracker options for this particular case, but then it becomes much
- less obvious how real auto-tracker will access this mask with old
- style API.
+ Doing it in bundled version, as the upstream library needs to have some
+ of the recent development ported over from bundle to upstream.
- So seems we do need an accessor for such data, just matter of
- finding better place than frame accessor.
+ There should be no functional changes.
-commit faa069cb826892780356477cc10602390fecf06b
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed May 31 10:36:26 2017 +0200
+commit c4de0ccd5aede3bb90ac6ad1039a83c2260fbefd
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Tue Feb 23 16:43:01 2021 +0100
- Tests: Tweak epsilon to avoid what looks a false-positive failure
+ Avoid use of LOG(INFO) in solver
+
+ Usage of LOG(INFO) actually went against own guidelines in the
+ logging.h: the INFO is for messages which are to be printed
+ regardless of debug/verbosity settings.
-commit 7c84e45c1d330871477ba3516f57178e5b9d101f
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Wed May 31 10:15:43 2017 +0200
+commit 94d925131b08e23cea7fdf735923e24b78a8c7fd
+Author: Campbell Barton <ideasman42 at gmail.com>
+Date: Fri Feb 5 16:23:34 2021 +1100
- CMake: Fix mistake in closing branch
+ Cleanup: correct spelling in comments
-commit cb769a0d319a8c95948153d78a4c3378a0142ece
-Author: Sergey Sharybin <sergey.vfx at gmail.com>
-Date: Thu Jul 21 12:52:33 2016 +0200
+commit db0c8dbea19d684899dffcb46a2e79c12dc91e6b
+Author: Sergey Sharybin <sergey at blender.org>
+Date: Tue Dec 1 14:52:08 2020 +0100
- Set of fixes for MSVC215
+ Tweak default logging verbosity level
+
+ Log to verbosity level 1 rather than INFO severity.
- - Move GLOG/GFLAGS defines to a more global scope,
- this way ANY of our own libraries will use proper
- declspec.
+ Avoids a lot of overhead coming from construction of the INFO stream
+ and improves performance and threadability of code which uses logging.
- - Compile png/zlib/openexif on Windows as well since
-
@@ Diff output truncated at 10240 characters. @@
More information about the Bf-blender-cvs
mailing list