Commit Graph

  • 47ca334d41 fix(python): handle byte-swapped NumPy inputs on 5.x Vladimir Iglovikov 2026-10-29 21:39:48 +05:45
  • c7c32d37eb imgproc: fix heap overflow in Laplacian for CV_16F Shiven Solanki 2026-10-29 23:24:14 +08:00
  • 5584875dba Merge pull request #30532 from gksks1029/fix/tracker-vit-roi-crop-4x Abhishek Gola 2026-10-29 19:41:46 +05:30
  • 2f5613fd94 Merge pull request #30559 from uwezkhan:flann-kmeans-load-roots uwezkhan 2026-10-29 18:18:36 +05:30
  • 8143864783 Merge pull request #30539 from sshivensolanki/fix-checkrange-fp8 Abhishek Gola 2026-10-29 18:05:20 +05:30
  • 5fc1aaf807 Merge pull request #30537 from sshivensolanki/fix-reduce-arg-new-depths Abhishek Gola 2026-10-29 18:03:13 +05:30
  • dd7163bdce Merge pull request #30092 from gupengzu/calcHist-parallel Abhishek Gola 2026-10-29 16:38:10 +05:30
  • f5efb49add Fix opencv_world losing MLAS's externally-assembled ARM64 kernel objects Prasadayus 2026-10-29 16:36:07 +05:30
  • e3181e617e core: rename is_fp16 to is_small_float in cv::norm Shiven Solanki 2026-10-29 16:37:25 +08:00
  • c55f0e9418 Merge pull request #30445 from 3nK1i/fix/convertmaps-empty-output Abhishek Gola 2026-10-29 13:39:54 +05:30
  • 099dde0419 Merge pull request #30452 from pbkx/fix-seamless-cloning-mask-preservation-4x Abhishek Gola 2026-10-29 13:37:55 +05:30
  • 981ff0ff05 Merge pull request #30508 from mh-masum6875/docs-python-manual-signatures Abhishek Gola 2026-10-29 13:36:36 +05:30
  • fa545b8b32 Merge pull request #30512 from Kaustubh1235/dnn-quantize-int16 Abhishek Gola 2026-10-29 13:35:38 +05:30
  • 57a97025ef Merge pull request #30498 from pbkx/fix-tonemap-reinhard-intensity-4x Abhishek Gola 2026-10-29 12:50:43 +05:30
  • 2d10bc004c dnn: run the patch-embedding GEMM once per image so MLAS threads it, and index copyChannelBlocks pieces directly Om Navin Rope 2026-10-28 20:08:53 +05:30
  • 2bf1f0cc46 dnn: take the patch-embedding im2col buffer from the planner, skip empty convs in forward() Om Navin Rope 2026-10-26 14:26:29 +05:30
  • 6d82cef4cf dnn: handle zero-batch tensors in block-layout Split, Concat and the blocked conv kernels Om Navin Rope 2026-10-23 14:41:17 +05:30
  • 7c7e14e48d dnn: drop externalArgs from the LayerScale fold, useCounts() already counts graph outputs Om Navin Rope 2026-10-21 14:03:02 +05:30
  • 0c943d143a dnn: run only ViT patch embeddings on the plain input Om Navin Rope 2026-10-14 20:17:31 +05:30
  • 1b1eeb401d dnn: keep core's copy policy, use the MLAS im2col conv only for constant weights, rename its helpers Om Navin Rope 2026-10-12 14:27:04 +05:30
  • ba06eb7cb8 dnn: keep block layout through Split and copy block-layout Concat in parallel Om Navin Rope 2026-10-11 15:41:36 +05:30
  • 35b2e73bb7 dnn: faster convs for few input channels and strided 3x3, and stop 1x9/9x1 kernels taking the 3x3 path Om Navin Rope 2026-10-11 15:41:36 +05:30
  • 8522306579 dnn: fold a per-channel Mul/Add after a constant MatMul (ViT LayerScale) into the Gemm weights and bias Om Navin Rope 2026-10-11 15:41:36 +05:30
  • 36836bb4cc Merge pull request #30313 from Prasadayus:dnn-block-layout-optimization Prasad Ayush Kumar 2026-10-29 11:11:09 +05:30
  • 3ed7253804 Merge pull request #30307 from vrabaud:ubsan Vincent Rabaud 2026-10-29 06:38:32 +01:00
  • c415e4fcef Merge pull request #30526 from sshivensolanki:fix-minmax-fp8 Shiven Solanki 2026-10-29 13:35:30 +08:00
  • b5768a806d Merge pull request #30128 from asmorkalov:as/kleidicv_26.09 Alexander Smorkalov 2026-10-29 08:08:01 +03:00
  • 684f8c445e core: support fp8 depths in cv::norm Shiven Solanki 2026-10-29 12:20:06 +08:00
  • 73b5a78e6a core: compare toggled fp8 bytes in checkRange Shiven Solanki 2026-10-29 11:31:06 +08:00
  • 18520b5674 core: compare toggled fp8 bytes in reduceArgMax and reduceArgMin Shiven Solanki 2026-10-29 11:09:32 +08:00
  • 50cfa93293 dnn(torch): validate nn.Padding dimension and pad bounds Nathan Keys 2026-10-28 22:02:46 -05:00
  • e52dcc2516 Merge pull request #30534 from sshivensolanki:fix-lut-fp8 Alexander Smorkalov 2026-10-28 21:36:30 +03:00
  • 2b1593eecc Merge pull request #30520 from intel-staging:dev/tizmajlo/icv_wa_rev Alexander Smorkalov 2026-10-28 21:33:10 +03:00
  • 3e93353b5b Merge pull request #30530 from uwezkhan:sce-weights-shape-check uwezkhan 2026-10-29 00:00:37 +05:30
  • 02dcf56058 Merge pull request #30276 from SavyaSanchi-Sharma:opti/consts Savya Sanchi Sharma 2026-10-28 23:58:47 +05:30
  • 9685cfc917 features: match XFeat's resize interpolation in letterbox test CheesyWannabe 2026-10-29 01:10:03 +08:00
  • 77d9a8e69c python: align setMouseCallback argument names with its keywords Maidul Hasan Masum 2026-10-28 14:13:46 +00:00
  • 895718ec0d Merge pull request #30283 from Prasadayus/mmap_external_initializers Abhishek Gola 2026-10-28 19:11:25 +05:30
  • a83d9971b0 core: support CV_8S and CV_32S in cv::reduce pranayr710 2026-10-28 16:46:51 +05:30
  • 3f7c2f7077 python: limit setAllowNonConventionalDataTypes() to outputs, fix short ml_dtypes operands, document it Abhishek Gola 2026-10-28 18:14:23 +05:30
  • ad833bee96 Merge pull request #30549 from mh-masum6875:docs-python-signatures-5x Maidul Hasan Masum 2026-10-28 19:28:52 +08:00
  • 950d4fe0ae features: use EXPECT_THROW for invalid LightGlue keypoint normalization CheesyWannabe 2026-10-28 19:27:46 +08:00
  • 21a2f2b8df Merge pull request #30131 from asmorkalov:as/blobFromImage_bitexact_flag Alexander Smorkalov 2026-10-28 13:04:00 +03:00
  • 01499d5150 cmake: enable clang-cl build on Windows ARM64 with KleidiCV Pratham Kumar 2026-10-28 13:30:02 +05:30
  • 2001806ed2 features: add max-side keypoint normalization to LightGlueMatcher CheesyWannabe 2026-10-28 10:51:37 +08:00
  • 96f8c0761e Merge pull request #30550 from asmorkalov:as/ts_fp8 Alexander Smorkalov 2026-10-27 21:16:35 +03:00
  • dd44ca21aa Merge pull request #30515 from asmorkalov:as/win_ffmpeg_9.0.2 Alexander Smorkalov 2026-10-27 18:17:14 +03:00
  • fc66429549 features: fix cv::XFeat letterbox inverse mapping CheesyWannabe 2026-10-27 21:29:51 +08:00
  • f2efb42906 core: use TEST_P for the 8-byte copyTo/setTo mask test Shiven Solanki 2026-10-27 18:03:23 +08:00
  • 8bc29c85a6 features: fix cv::XFeat letterbox inverse mapping CheesyWannabe 2026-10-27 17:16:46 +08:00
  • c5451c08b6 Extend TS module ENUM to print FP8 data type names properly. Alexander Smorkalov 2026-10-27 12:04:10 +03:00
  • ba2925935b Merge pull request #30479 from Prasadayus/fix/lapack-hal-decline-transpose Abhishek Gola 2026-10-27 11:59:49 +05:30
  • 9ebb2b91a3 Merge remote-tracking branch 'origin/5.x' into dnn-quantize-int16-merge Kaustubh Arora 2026-10-27 02:09:13 +05:30
  • 3c3e508949 core: support fp8 depths in checkRange Shiven Solanki 2026-10-27 02:55:18 +08:00
  • cd3dc424fb Merge pull request #30463 from sshivensolanki:fix-countnonzero-fp8 Shiven Solanki 2026-10-27 02:23:59 +08:00
  • 67c2e2c3ff core: support CV_Bool and fp8 depths in reduceArgMax and reduceArgMin Shiven Solanki 2026-10-27 02:16:24 +08:00
  • fca52f6284 core: support fp8 lookup tables in LUT Shiven Solanki 2026-10-27 01:06:00 +08:00
  • f7ca3a382a Merge pull request #30416 from Prasadayus:dnn-int8-layer-support Prasad Ayush Kumar 2026-10-26 21:22:12 +05:30
  • 58e52e9d3c video: isolate TrackerVit crops from parent images Shih Kai Siang 2026-10-26 23:17:43 +08:00
  • fa3037533e Merge pull request #30524 from Kumataro:doc/avif_support_16bit_4.x Alexander Smorkalov 2026-10-26 16:20:58 +03:00
  • d9b582d464 Merge pull request #30527 from Laibaqi:docs-fix-android-typos Alexander Smorkalov 2026-10-26 16:20:26 +03:00
  • 91ca181f19 docs: fix typos in Android development tutorial Laiba Irfan 2026-10-26 17:45:06 +05:00
  • 1a97289f43 doc(imgcodecs): update 16-bit support and opaque alpha channel behavior for AVIF(4.x) - Reading and writing 16-bit AVIF images requires libavif v1.4.0+ - Note that if the alpha plane contains only fully opaque pixels, the destination saves only color channels due to a libavif limitation. Kumataro 2026-10-26 20:42:26 +09:00
  • 4d7924ebb5 Merge pull request #30507 from tushar152006:fix-onnx-bool-tensor-parsing Alexander Smorkalov 2026-10-26 14:30:05 +03:00
  • 3a100454f2 Merge pull request #30519 from Kumataro:doc/avif_support_16bit Alexander Smorkalov 2026-10-26 14:13:27 +03:00
  • 8ad70d0f30 Merge pull request #30302 from varun-jaiswal17/dnn-conv-task-sizing Abhishek Gola 2026-10-26 16:10:22 +05:30
  • bd7da3fa78 Merge pull request #30510 from uwezkhan:crop-and-resize-bounds Alexander Smorkalov 2026-10-26 13:30:45 +03:00
  • 87e9907bb7 icv: a workaround for the issue in ICV package on Windows is reverted Izmajlov, Timur 2026-10-23 08:55:55 -07:00
  • ba021a17a8 doc(imgcodecs): update 16-bit support and opaque alpha channel behavior for AVIF Kumataro 2026-10-26 19:09:42 +09:00
  • 9171eaf2fa Merge pull request #30242 from pratham-mcw/inpaint_opt Abhishek Gola 2026-10-26 14:18:55 +05:30
  • 2b7cd24889 Merge pull request #30514 from Kaustubh1235:dnn-conformance-stale-qlinear Alexander Smorkalov 2026-10-26 11:15:52 +03:00
  • 8272f64611 Merge pull request #30516 from velonica0:okriscv/imgproc-rvv-hsv2bgr-vmerge Alexander Smorkalov 2026-10-26 10:22:43 +03:00
  • 46dba3df6e Merge pull request #30506 from sshivensolanki:fix-sort-new-depths Shiven Solanki 2026-10-26 15:05:01 +08:00
  • f98c0a46fa Merge pull request #30499 from Rishiii57:add-avif-alpha-quality-30451 Rishi Kumar 2026-10-26 11:31:16 +05:30
  • 30b34666fb imgproc: riscv-rvv: replace sector table gathers in Hxx2BGR with vmerge velonica0 2026-10-25 18:52:25 -07:00
  • 3bea7d2b1c dnn: enable the QLinearMatMul dtype variants and blocked asymmetric QuantizeLinear conformance cases Kaustubh Arora 2026-10-25 23:14:45 +05:30
  • abd21b8d54 dnn: validate boxes shape and follow TF extrapolation in CropAndResize Uwez Khan 2026-10-24 13:39:56 +05:30
  • eb604b9a2b Merge pull request #30513 from asmorkalov:as/ffmpeg_8 Alexander Smorkalov 2026-10-25 18:34:48 +03:00
  • 991ac72e2f Port FFMpeg 8.0 support from 29533 Alexander Smorkalov 2026-10-25 17:49:54 +03:00
  • ba8f8681c5 dnn: add the int16 QuantizeLinear/DequantizeLinear cases to the OpenVINO filter Kaustubh Arora 2026-10-25 19:28:53 +05:30
  • a25e124d2b dnn: enable test_quantizelinear_blocked_symmetric Kaustubh Arora 2026-10-25 18:04:02 +05:30
  • d16c6529a2 dnn(onnx): support boolean tensors stored in int32_data Tushar Burla 2026-10-24 22:37:15 +05:30
  • 7dc486f74c Merge pull request #30497 from Kumataro:fix_norm_calculation_and_extend_16bit_tests_4.x Alexander Smorkalov 2026-10-25 10:56:25 +03:00
  • f365616a02 Merge pull request #30503 from lazerg/fix:issue-30501-gathernd-empty-indices Lazizbek Ergashev 2026-10-25 12:11:04 +05:00
  • 92c551df0e dnn: support int16/uint16 in QuantizeLinear and DequantizeLinear Kaustubh Arora 2026-10-25 03:49:39 +05:30
  • 32f9288034 python: add documentation signatures for manually wrapped highgui functions Maidul Hasan Masum 2026-10-25 02:16:36 +08:00
  • 56dabe2b19 Merge pull request #30289 from varun-jaiswal17:dnn-int8-cpu-perf Varun Jaiswal 2026-10-24 15:01:46 +05:30
  • 048f050d5d Merge pull request #30475 from asmorkalov:as/tanh_clamp_amrv7a Alexander Smorkalov 2026-10-24 12:03:58 +03:00
  • 148d007185 Parallelize calcHist using thread-local histograms gupengzu 2026-10-24 11:17:34 +08:00
  • 4187e46d27 preserve TonemapReinhard intensity during processing pbkx 2026-10-23 19:11:54 -07:00
  • b070b228e4 imgcodecs(avif): fix norm calculation and extend 16-bit tests Kumataro 2026-10-24 06:53:33 +09:00
  • a1e798a74e use test_p for parameterized test Prasadayus 2026-10-24 01:07:01 +05:30
  • e1d90775be core: fix convertTo() losing values above int32 when converting from FP to 64-bit int Tushar Burla 2026-10-23 04:25:47 +05:30
  • 4058787c82 Merge pull request #30407 from asmorkalov:as/ocl_fp64 Alexander Smorkalov 2026-10-23 18:34:50 +03:00
  • 3747f5a3f1 core: restore input matrix when LAPACK HAL declines Prasadayus 2026-10-23 18:10:21 +05:30
  • 3f12ad88ab Merge pull request #30476 from asmorkalov:as/pre_5.1.0 Alexander Smorkalov 2026-10-23 15:31:14 +03:00
  • d23780fb56 Merge pull request #30431 from Kumataro:trial/avif_update_test_for_each_bitdepth Kumataro 2026-10-23 21:02:23 +09:00
  • 54fbb6bd86 Pre-release 5.1.0 versions update Alexander Smorkalov 2026-10-23 14:28:38 +03:00
  • 9b2edb0ed0 resolved comments call the depthwise HAL once; the RVV hook runs its own parallel_for_ vrooomy 2026-10-23 16:43:16 +05:30
  • b423992c26 imgproc: avoid narrowing warnings in convertMaps tests Enki 2026-10-23 10:37:18 +00:00