[Config Support]: dewarp 360 camera using intel's QSV #15920
Answered
by
gaynetdinov
gaynetdinov
asked this question in
Config Support
-
Describe the problem you are havingHey folks! After you helped me to fix my QSV config in previous thread #15861 I would like to finally try to dewarp my 360 reolink fe-p camera. I have a command that works on my mac when I connect to go2rtc stream of my frigate instance. I can see the output exactly how I want it to be Then it seems like I figured out all the quirks regarding QSV hardware acceleration and this command works in the frigate docker instance no errors about DTS or bad filters, I will attach of this output to the go2rtc log output section below. But when I put this ffmpeg arguments to frigate config, it does not work. Version0.15-beta4 Frigate config fileffmpeg:
hwaccel_args: preset-intel-qsv-h264
global_args: -hide_banner -loglevel warning -threads 4 -bsf:v dump_extra
detectors:
ov:
type: openvino
device: GPU
model:
path: plus://
model:
path: plus://
go2rtc:
streams:
carport:
- ffmpeg:https://192.168.2.184/flv?port=1935&app=bcs&stream=channel0_main.bcs&user=admin&password=pass#video=copy#audio=copy#audio=opus
carport_sub:
- ffmpeg:https://192.168.2.184/flv?port=1935&app=bcs&stream=channel0_ext.bcs&user=admin&password=pass
hall:
- ffmpeg:https://192.168.2.204/flv?port=1935&app=bcs&stream=channel0_main.bcs&user=admin&password=pass#video=copy#audio=copy#audio=opus
hall_sub:
- ffmpeg:https://192.168.2.204/flv?port=1935&app=bcs&stream=channel0_sub.bcs&user=admin&password=pass
doorbell:
- ffmpeg:rtsp://admin:[email protected]/Streaming/Channels/101/
doorbell_sub:
- ffmpeg:rtsp://admin:[email protected]/Streaming/Channels/102/
cameras:
doorbell:
record:
enabled: true
retain:
days: 3
mode: motion
objects:
track:
- amazon
- dhl
- usps
- ups
- fedex
- gls
- dpd
ffmpeg:
inputs:
- path: rtsp://127.0.0.1:8554/doorbell
input_args: preset-rtsp-restream
roles:
- audio
- record
- path: rtsp://127.0.0.1:8554/doorbell_sub
input_args: preset-rtsp-restream
roles:
- detect
motion:
mask: 0.673,0.767,0.974,0.764,0.974,0.826,0.674,0.825
hall:
detect:
width: 1024
height: 1024
record:
enabled: true
retain:
days: 3
mode: motion
objects:
track:
- person
- cat
- face
ffmpeg:
inputs:
- path: rtsp://127.0.0.1:8554/hall
input_args: preset-rtsp-restream
roles:
- audio
- record
- path: rtsp://127.0.0.1:8554/hall_sub
input_args: -rtsp_transport tcp -timeout 30000000
roles:
- detect
#output_args:
# record: -vf "hwdownload,format=nv12,format=yuv420p,v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy" -f segment -segment_time 10 -segment_format mp4
#record: -vf hwdownload,format=nv12,format=yuv420p,v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy -f segment -segment_time 10 -segment_format mp4 -reset_timestamps 1 -strftime 1
#detect: -vf "hwdownload,format=nv12,format=yuv420p,v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy" -r 10 -b:v 3M -maxrate 3M -bufsize 6M -c:v h264_qsv -f rawvideo -flags +cgop
#record: -vf "v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy" -f segment -segment_time 10 -segment_format mp4 -reset_timestamps 1 -strftime 1 -c:v h264_qsv
#detect: -vf "v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy" -f segment -segment_time 10 -segment_format mp4 -reset_timestamps 1 -strftime 1
# -vf "hwdownload,format=nv12,v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy" -r 10 -b:v 3M -maxrate 3M -bufsize 6M -c:v h264_qsv -flags +cgop
carport:
objects:
track:
- person
- cat
- car
- dog
- bicycle
- fox
- raccoon
- squirrel
- rabbit
- umbrella
- motorcycle
- license_plate
ffmpeg:
inputs:
- path: rtsp://127.0.0.1:8554/carport
input_args: preset-rtsp-restream
roles:
- audio
- record
- path: rtsp://127.0.0.1:8554/carport_sub
input_args: preset-rtsp-restream
roles:
- detect
motion:
mask:
- 0.708,0,0.707,0.079,0.287,0.08,0.285,0
- 0.187,0.379,0.193,0.452,0.124,0.495,0.1,0.192,0.35,0.192,0.355,0.281
threshold: 40
contour_area: 20
improve_contrast: true
zones:
Carport:
coordinates:
0.132,0.497,0.197,0.462,0.185,0.386,0.269,0.323,0.355,0.292,1,0.736,0.997,0.997,0.215,0.993
loitering_time: 0
inertia: 3
review:
alerts: {}
detections: {}
record:
enabled: true
retain:
days: 3
mode: all
snapshots:
enabled: true
bounding_box: true
timestamp: true
crop: true
quality: 100
retain:
default: 60
review:
# Optional: alerts configuration
alerts:
# Optional: labels that qualify as an alert (default: shown below)
labels:
- car
- person
- cat
- fox
- raccoon
- squirrel
- rabbit
version: 0.15-4
semantic_search:
enabled: true
reindex: true
model_size: smallRelevant Frigate log outputfrigate | 2025-01-09 17:34:48.371493665 [2025-01-09 17:34:48] ffmpeg.hall.audio_record ERROR : Impossible to convert between the formats supported by the filter 'graph 0 input from stream 0:0' and the filter 'auto_scale_0'
frigate | 2025-01-09 17:34:48.378154763 [2025-01-09 17:34:48] ffmpeg.hall.audio_record ERROR : [vf#0:0 @ 0x63db01909740] Error reinitializing filters!
frigate | 2025-01-09 17:34:48.378443286 [2025-01-09 17:34:48] ffmpeg.hall.audio_record ERROR : [vf#0:0 @ 0x63db01909740] Task finished with error code: -38 (Function not implemented)
frigate | 2025-01-09 17:34:48.378627328 [2025-01-09 17:34:48] ffmpeg.hall.audio_record ERROR : [vf#0:0 @ 0x63db01909740] Terminating thread with return code -38 (Function not implemented)
frigate | 2025-01-09 17:34:48.378629700 [2025-01-09 17:34:48] ffmpeg.hall.audio_record ERROR : [vost#0:0/libx264 @ 0x63db0190f3c0] Could not open encoder before EOFRelevant go2rtc log outputroot@5c8409d55e23:/usr/lib/ffmpeg/7.0/bin# ./ffmpeg -v verbose -hwaccel qsv -bsf:v dump_extra -i rtsp://localhost:8554/hall_sub -vf "hwdownload,format=nv12,v360=fisheye:flat:id_fov=180:d_fov=90:roll=30:pitch=20:rorder=rpy" -r 10 -b:v 3M -maxrate 3M -bufsize 6M -c:v h264_qsv -flags +cgop -f null -
ffmpeg version n7.0.2-18-g3e6cec1286-20240919 Copyright (c) 2000-2024 the FFmpeg developers
built with gcc 14.2.0 (crosstool-NG 1.26.0.106_ed12fa6)
configuration: --prefix=/ffbuild/prefix --pkg-config-flags=--static --pkg-config=pkg-config --cross-prefix=x86_64-ffbuild-linux-gnu- --arch=x86_64 --target-os=linux --enable-gpl --enable-version3 --disable-debug --enable-iconv --enable-zlib --enable-libfreetype --enable-libfribidi --enable-gmp --enable-libxml2 --enable-openssl --enable-lzma --enable-fontconfig --enable-libharfbuzz --enable-libvorbis --enable-opencl --enable-libpulse --enable-libvmaf --enable-libxcb --enable-xlib --enable-amf --enable-libaom --enable-libaribb24 --enable-avisynth --enable-chromaprint --enable-libdav1d --enable-libdavs2 --enable-libdvdread --enable-libdvdnav --disable-libfdk-aac --enable-ffnvcodec --enable-cuda-llvm --enable-frei0r --enable-libgme --enable-libkvazaar --enable-libaribcaption --enable-libass --enable-libbluray --enable-libjxl --enable-libmp3lame --enable-libopus --enable-librist --enable-libssh --enable-libtheora --enable-libvpx --enable-libwebp --enable-libzmq --enable-lv2 --enable-libvpl --enable-openal --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenh264 --enable-libopenjpeg --enable-libopenmpt --enable-librav1e --enable-librubberband --disable-schannel --enable-sdl2 --enable-libsoxr --enable-libsrt --enable-libsvtav1 --enable-libtwolame --enable-libuavs3d --enable-libdrm --enable-vaapi --enable-libvidstab --enable-vulkan --enable-libshaderc --enable-libplacebo --enable-libx264 --enable-libx265 --enable-libxavs2 --enable-libxvid --enable-libzimg --enable-libzvbi --extra-cflags=-DLIBTWOLAME_STATIC --extra-cxxflags= --extra-libs='-ldl -lgomp' --extra-ldflags=-pthread --extra-ldexeflags=-pie --cc=x86_64-ffbuild-linux-gnu-gcc --cxx=x86_64-ffbuild-linux-gnu-g++ --ar=x86_64-ffbuild-linux-gnu-gcc-ar --ranlib=x86_64-ffbuild-linux-gnu-gcc-ranlib --nm=x86_64-ffbuild-linux-gnu-gcc-nm --extra-version=20240919
libavutil 59. 8.100 / 59. 8.100
libavcodec 61. 3.100 / 61. 3.100
libavformat 61. 1.100 / 61. 1.100
libavdevice 61. 1.100 / 61. 1.100
libavfilter 10. 1.100 / 10. 1.100
libswscale 8. 1.100 / 8. 1.100
libswresample 5. 1.100 / 5. 1.100
libpostproc 58. 1.100 / 58. 1.100
[tcp @ 0x61c30b6b3880] Starting connection attempt to ::1 port 8554
[tcp @ 0x61c30b6b3880] Successfully connected to ::1 port 8554
[rtsp @ 0x61c30b6b2e80] SDP:
v=0
o=- 1 1 IN IP4 0.0.0.0
s=go2rtc/1.9.2
c=IN IP4 0.0.0.0
t=0 0
m=video 0 RTP/AVP 96
a=rtpmap:96 H264/90000
a=fmtp:96 packetization-mode=1; sprop-parameter-sets=Z2QAM6wVFKBAAgZA,aO48sA==; profile-level-id=640033
a=control:trackID=0
m=audio 0 RTP/AVP 97
a=rtpmap:97 MPEG4-GENERIC/16000
a=fmtp:97 profile-level-id=1;mode=AAC-hbr;sizelength=13;indexlength=3;indexdeltalength=3; config=1408
a=control:trackID=1
[rtsp @ 0x61c30b6b2e80] method SETUP failed: 461 Unsupported transport
[rtsp @ 0x61c30b6b2e80] setting jitter buffer size to 0
Last message repeated 1 times
[h264 @ 0x61c30b6b4740] Reinit context to 1024x1024, pix_fmt: yuv420p
[rtsp @ 0x61c30b6b2e80] max_analyze_duration 5000000 reached at 5056000 microseconds st:1
[vist#0:0/h264 @ 0x61c30b6e09c0] WARNING: defaulting hwaccel_output_format to qsv for compatibility with old commandlines. This behaviour is DEPRECATED and will be removed in the future. Please explicitly set "-hwaccel_output_format qsv".
Selecting decoder 'h264_qsv' because of requested hwaccel method qsv
Input #0, rtsp, from 'rtsp://localhost:8554/hall_sub':
Metadata:
title : go2rtc/1.9.2
Duration: N/A, start: 0.000000, bitrate: N/A
Stream #0:0: Video: h264 (High), 1 reference frame, yuv420p(progressive), 1024x1024, 25 tbr, 90k tbn
Stream #0:1: Audio: aac (LC), 16000 Hz, mono, fltp
[out#0/null @ 0x61c30b6f4000] No explicit maps, mapping streams automatically...
[vost#0:0/h264_qsv @ 0x61c30b6e1400] Created video stream from input stream 0:0
[AVHWDeviceContext @ 0x61c30b721580] Trying to use DRM render node for device 0, with matching kernel driver (i915).
[AVHWDeviceContext @ 0x61c30b721580] libva: VA-API version 1.22.0
[AVHWDeviceContext @ 0x61c30b721580] libva: User requested driver 'iHD'
[AVHWDeviceContext @ 0x61c30b721580] libva: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
[AVHWDeviceContext @ 0x61c30b721580] libva: Found init function __vaDriverInit_1_22
[AVHWDeviceContext @ 0x61c30b721580] libva: va_openDriver() returns 0
[AVHWDeviceContext @ 0x61c30b721580] Initialised VAAPI connection: version 1.22
[AVHWDeviceContext @ 0x61c30b721580] VAAPI driver: Intel iHD driver for Intel(R) Gen Graphics - 24.3.3 ().
[AVHWDeviceContext @ 0x61c30b721580] Driver not found in known nonstandard list, using standard behaviour.
[AVHWDeviceContext @ 0x61c30b6f2080] Use Intel(R) oneVPL to create MFX session, API version is 2.13, the required implementation version is 1.3
libva info: VA-API version 1.22.0
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
libva info: Found init function __vaDriverInit_1_22
libva info: va_openDriver() returns 0
libva info: VA-API version 1.22.0
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
libva info: Found init function __vaDriverInit_1_22
libva info: va_openDriver() returns 0
[AVHWDeviceContext @ 0x61c30b6f2080] Initialize MFX session: implementation version is 2.11
[h264_mp4toannexb @ 0x61c30bd03200] The input looks like it is Annex B already
[aost#0:1/pcm_s16le @ 0x61c30bd04200] Created audio stream from input stream 0:1
Stream mapping:
Stream #0:0 -> #0:0 (h264 (h264_qsv) -> h264 (h264_qsv))
Stream #0:1 -> #0:1 (aac (native) -> pcm_s16le (native))
[vost#0:0/h264_qsv @ 0x61c30b6e1400] Starting thread...
[aost#0:1/pcm_s16le @ 0x61c30bd04200] Starting thread...
[vf#0:0 @ 0x61c30b74c240] Starting thread...
[af#0:1 @ 0x61c30ba09b00] Starting thread...
[vist#0:0/h264 @ 0x61c30b6e09c0] [dec:h264_qsv @ 0x61c30b843b00] Starting thread...
[aist#0:1/aac @ 0x61c30b720cc0] [dec:aac @ 0x61c30ba6dd00] Starting thread...
[in#0/rtsp @ 0x61c30b6b2b40] Starting thread...
Press [q] to stop, [?] for help
[AVHWDeviceContext @ 0x7d4abc002d40] VAAPI driver: Intel iHD driver for Intel(R) Gen Graphics - 24.3.3 ().
[AVHWDeviceContext @ 0x7d4abc002d40] Driver not found in known nonstandard list, using standard behaviour.
[h264_qsv @ 0x61c30b8444c0] Decoder: output is video memory surface
[h264_qsv @ 0x61c30b8444c0] Use Intel(R) oneVPL to create MFX session with the specified MFX loader
[graph_1_in_0:1 @ 0x7d4ac4003080] tb:1/16000 samplefmt:fltp samplerate:16000 chlayout:mono
[format_out_0_1 @ 0x7d4ac4004280] auto-inserting filter 'auto_aresample_0' between the filter 'Parsed_anull_0' and the filter 'format_out_0_1'
[auto_aresample_0 @ 0x7d4ac4005d80] ch:1 chl:mono fmt:fltp r:16000Hz -> ch:1 chl:mono fmt:s16 r:16000Hz
[h264_qsv @ 0x61c30b8444c0] More data is required to decode header
Last message repeated 34 times
[AVHWDeviceContext @ 0x7d4abc0144c0] VAAPI driver: Intel iHD driver for Intel(R) Gen Graphics - 24.3.3 ().
[AVHWDeviceContext @ 0x7d4abc0144c0] Driver not found in known nonstandard list, using standard behaviour.
[h264_qsv @ 0x61c30b8444c0] Decoder: output is video memory surface
[h264_qsv @ 0x61c30b8444c0] Use Intel(R) oneVPL to create MFX session with the specified MFX loader
[graph 0 input from stream 0:0 @ 0x7d4acc005a40] w:1024 h:1024 pixfmt:qsv tb:1/90000 fr:25/1 sar:0/1 csp:unknown range:tv
[auto_scale_0 @ 0x7d4acc001a40] w:iw h:ih flags:'' interl:0
[Parsed_v360_2 @ 0x7d4acc003d80] auto-inserting filter 'auto_scale_0' between the filter 'Parsed_format_1' and the filter 'Parsed_v360_2'
[auto_scale_1 @ 0x7d4acc017e80] w:iw h:ih flags:'' interl:0
[format @ 0x7d4acc0074c0] auto-inserting filter 'auto_scale_1' between the filter 'Parsed_v360_2' and the filter 'format'
[auto_scale_0 @ 0x7d4acc001a40] w:1024 h:1024 fmt:nv12 csp:unknown range:tv sar:0/1 -> w:1024 h:1024 fmt:yuv420p csp:unknown range:tv sar:0/1 flags:0x00000004
[auto_scale_1 @ 0x7d4acc017e80] w:2048 h:1024 fmt:yuv420p csp:unknown range:tv sar:0/1 -> w:2048 h:1024 fmt:nv12 csp:unknown range:tv sar:0/1 flags:0x00000004
[graph 0 input from stream 0:0 @ 0x7d4acc005a40] video frame properties congruent with link at pts_time: 3.519
[AVHWFramesContext @ 0x7d4abc015180] Use Intel(R) oneVPL to create MFX session, API version is 2.13, the required implementation version is 2.11
[AVHWFramesContext @ 0x7d4abc015180] Initialize MFX session: implementation version is 2.11
[vost#0:0/h264_qsv @ 0x61c30b6e1400] *** 35 dup!
[h264_qsv @ 0x61c30b6efcc0] Using device qsv0 (type qsv) with h264_qsv encoder.
[h264_qsv @ 0x61c30b6efcc0] Encoder: input is system memory surface
[h264_qsv @ 0x61c30b6efcc0] Use Intel(R) oneVPL to create MFX session with the specified MFX loader
[h264_qsv @ 0x61c30b6efcc0] Using the constant bitrate (CBR) ratecontrol method
[h264_qsv @ 0x61c30b6efcc0] profile: avc high; level: 40
[h264_qsv @ 0x61c30b6efcc0] GopPicSize: 256; GopRefDist: 3; GopOptFlag: closed; IdrInterval: 0
[h264_qsv @ 0x61c30b6efcc0] TargetUsage: 4; RateControlMethod: CBR
[h264_qsv @ 0x61c30b6efcc0] BufferSizeInKB: 750; InitialDelayInKB: 562; TargetKbps: 3000; MaxKbps: 3000; BRCParamMultiplier: 1
[h264_qsv @ 0x61c30b6efcc0] NumSlice: 1; NumRefFrame: 2
[h264_qsv @ 0x61c30b6efcc0] RateDistortionOpt: OFF
[h264_qsv @ 0x61c30b6efcc0] RecoveryPointSEI: OFF
[h264_qsv @ 0x61c30b6efcc0] VDENC: OFF
[h264_qsv @ 0x61c30b6efcc0] Entropy coding: CABAC; MaxDecFrameBuffering: 2
[h264_qsv @ 0x61c30b6efcc0] NalHrdConformance: ON; SingleSeiNalUnit: ON; VuiVclHrdParameters: OFF VuiNalHrdParameters: ON
[h264_qsv @ 0x61c30b6efcc0] FrameRateExtD: 1; FrameRateExtN: 10
[h264_qsv @ 0x61c30b6efcc0] IntRefType: 0; IntRefCycleSize: 0; IntRefQPDelta: 0
[h264_qsv @ 0x61c30b6efcc0] MaxFrameSize: 0; MaxSliceSize: 0
[h264_qsv @ 0x61c30b6efcc0] BitrateLimit: ON; MBBRC: OFF; ExtBRC: OFF
[h264_qsv @ 0x61c30b6efcc0] Trellis: auto
[h264_qsv @ 0x61c30b6efcc0] RepeatPPS: OFF; NumMbPerSlice: 0; LookAheadDS: 2x
[h264_qsv @ 0x61c30b6efcc0] AdaptiveI: OFF; AdaptiveB: OFF; BRefType:off
[h264_qsv @ 0x61c30b6efcc0] MinQPI: 0; MaxQPI: 0; MinQPP: 0; MaxQPP: 0; MinQPB: 0; MaxQPB: 0
[h264_qsv @ 0x61c30b6efcc0] DisableDeblockingIdc: 0
[h264_qsv @ 0x61c30b6efcc0] SkipFrame: no_skip
[h264_qsv @ 0x61c30b6efcc0] PRefType: default
[h264_qsv @ 0x61c30b6efcc0] TransformSkip: unknown
[h264_qsv @ 0x61c30b6efcc0] IntRefCycleDist: 0
[h264_qsv @ 0x61c30b6efcc0] LowDelayBRC: OFF
[h264_qsv @ 0x61c30b6efcc0] MaxFrameSizeI: 0; MaxFrameSizeP: 0
[h264_qsv @ 0x61c30b6efcc0] ScenarioInfo: 0
Output #0, null, to 'pipe:':
Metadata:
title : go2rtc/1.9.2
encoder : Lavf61.1.100
Stream #0:0: Video: h264, 1 reference frame, nv12(tv, progressive), 2048x1024, q=2-31, 3000 kb/s, 10 fps, 10 tbn
Metadata:
encoder : Lavc61.3.100 h264_qsv
Side data:
cpb: bitrate max/min/avg: 3000000/0/3000000 buffer size: 6000000 vbv_delay: N/A
Stream #0:1: Audio: pcm_s16le, 16000 Hz, mono, s16, 256 kb/s
Metadata:
encoder : Lavc61.3.100 pcm_s16le
[out#0/null @ 0x61c30b6f4000] Starting thread...
[in#0/rtsp @ 0x61c30b6b2b40] Terminating thread with return code 0 (success)0 speed=1.18x
[vist#0:0/h264 @ 0x61c30b6e09c0] [dec:h264_qsv @ 0x61c30b843b00] Decoder thread received EOF packet
[aist#0:1/aac @ 0x61c30b720cc0] [dec:aac @ 0x61c30ba6dd00] Decoder thread received EOF packet
[aist#0:1/aac @ 0x61c30b720cc0] [dec:aac @ 0x61c30ba6dd00] Decoder returned EOF, finishing
[aist#0:1/aac @ 0x61c30b720cc0] [dec:aac @ 0x61c30ba6dd00] Terminating thread with return code 0 (success)
[af#0:1 @ 0x61c30ba09b00] Filtergraph returned EOF, finishing
[af#0:1 @ 0x61c30ba09b00] All consumers returned EOF
[aost#0:1/pcm_s16le @ 0x61c30bd04200] Encoder thread received EOF
[aost#0:1/pcm_s16le @ 0x61c30bd04200] Terminating thread with return code 0 (success)
[af#0:1 @ 0x61c30ba09b00] Terminating thread with return code 0 (success)
[vist#0:0/h264 @ 0x61c30b6e09c0] [dec:h264_qsv @ 0x61c30b843b00] Decoder returned EOF, finishing
[vist#0:0/h264 @ 0x61c30b6e09c0] [dec:h264_qsv @ 0x61c30b843b00] Terminating thread with return code 0 (success)
[vf#0:0 @ 0x61c30b74c240] Filtergraph returned EOF, finishing
[vf#0:0 @ 0x61c30b74c240] All consumers returned EOF
[vost#0:0/h264_qsv @ 0x61c30b6e1400] Encoder thread received EOF
[vf#0:0 @ 0x61c30b74c240] Terminating thread with return code 0 (success)
[vost#0:0/h264_qsv @ 0x61c30b6e1400] Terminating thread with return code 0 (success)
[out#0/null @ 0x61c30b6f4000] All streams finished
[out#0/null @ 0x61c30b6f4000] Terminating thread with return code 0 (success)
[out#0/null @ 0x61c30b6f4000] Output file #0 (pipe:):
[out#0/null @ 0x61c30b6f4000] Output stream #0:0 (video): 196 frames encoded; 196 packets muxed (7165539 bytes);
[out#0/null @ 0x61c30b6f4000] Output stream #0:1 (audio): 306 frames encoded (313344 samples); 306 packets muxed (626688 bytes);
[out#0/null @ 0x61c30b6f4000] Total: 502 packets (7792227 bytes) muxed
[out#0/null @ 0x61c30b6f4000] video:6998KiB audio:612KiB subtitle:0KiB other streams:0KiB global headers:0KiB muxing overhead: unknown
frame= 196 fps= 13 q=13.0 Lsize=N/A time=00:00:19.40 bitrate=N/A dup=35 drop=0 speed=1.33x
[in#0/rtsp @ 0x61c30b6b2b40] Input file #0 (rtsp://localhost:8554/hall_sub):
[in#0/rtsp @ 0x61c30b6b2b40] Input stream #0:0 (video): 197 packets read (649605 bytes); 161 frames decoded; 0 decode errors;
[in#0/rtsp @ 0x61c30b6b2b40] Input stream #0:1 (audio): 306 packets read (156672 bytes); 306 frames decoded; 0 decode errors (313344 samples);
[in#0/rtsp @ 0x61c30b6b2b40] Total: 503 packets (806277 bytes) demuxedFrigate statsNo response Operating systemOther Linux Install methodDocker Compose docker-compose file or Docker CLI commandservices:
frigate:
container_name: frigate
privileged: true # this may not be necessary for all setups
restart: unless-stopped
image: ghcr.io/blakeblackshear/frigate:0.15.0-beta4
shm_size: "512mb" # update for your cameras based on calculation above
devices:
- /dev/dri/renderD128:/dev/dri/renderD128 # For intel hwaccel
volumes:
- /etc/localtime:/etc/localtime:ro
- /home/damirca/frigate/config:/config
- /home/damirca/frigate/storage:/media/frigate
- type: tmpfs # Optional: 1GB of memory, reduces SSD/SD Card wear
target: /tmp/cache
tmpfs:
size: 1000000000
ports:
- "8971:8971"
- "5000:5000" # Internal unauthenticated access. Expose carefully.
- "8554:8554" # RTSP feeds
- "8555:8555/tcp" # WebRTC over tcp
- "8555:8555/udp" # WebRTC over udp
environment:
- PLUS_API_KEY=blahObject DetectorOpenVino Screenshots of the Frigate UI's System metrics pagesNo response Any other information that may be helpfulNo response |
Beta Was this translation helpful? Give feedback.
Answered by
gaynetdinov
Jan 13, 2025
Replies: 1 comment 3 replies
-
|
If you are using a preset you don't have control over any video filters. Realistically, I would suggest doing the dewarping in go2rtc and then just pull that feed directly into frigate. That way dewarping only needs to be done once |
Beta Was this translation helpful? Give feedback.
3 replies
-
But if I want to have 2 streams out of 360 camera? I mean my current filter simply 'cuts out' one part of 360 picture. I thought go2rtc is exactly what I need for that so I could use local stream twice. Would that work? |
Beta Was this translation helpful? Give feedback.
-
|
I think I figured it out. Original resolution of the main stream of the camera is 2560x2560, however qsv cannot handle such resolution, so I have to scale down to 1920x1920 using Unfortunately my i5-12500 is having hard times processing 4 streams out of 2, Frigate reports around 50-60% CPU and 20-30% GPU load. So I decided to use substreams as source for 2 new dewarp 'cameras' and one main stream with good quality only for recording. With this setup I have detections and high quality recording (granted in 360 form) and Frigate reports not more than 20% CPU and around 5-15% GPU load handling in total of 5 cameras (2 real ones and 3 'cameras' made from a single fisheye camera). @NickM-27 I'd happy to know if there are any tips how to make this setup less CPU intensive. Would dedicated GPU help here? Thank you! |
Beta Was this translation helpful? Give feedback.
Answer selected by
gaynetdinov
-
|
I see the vpp_qsv at the end, that seems odd and most likely is not running on hardware since you have already done hardware download. Unfortunately, the dewarping being done on the CPU is likely the main source of CPU usage, as well as the moving between hardware and software memory |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I think I figured it out.
I was not aware that one could build streams on top of another streams within go2rtc config, I thought one could use one defined stream in go2rtc multiple times in
camerassetup. Also I was not aware of go2rtc'sexecfeature. With it it's much much simpler to apply sophisticated filters to the streams.Original resolution of the main stream of the camera is 2560x2560, however qsv cannot handle such resolution, so I have to scale down to 1920x1920 using
vpp_qsv=w=1920:h=1920. Then I usevpp_qsvonce again to have proper resolution of the output stream.