DeepStream With REST API Sever (Alpha)

DeepStream application supports runtime parameter configuration for certain components with the help of REST APIs. DeepStream application should run as a server for this feature.

Rest Server support is enabled for both x86 (DeepStream-6.2 release onwards) and Jetson (DeepStream-6.3 release onwards). REST Server library integrates the HTTP server using the “Civetweb” OSS library to process the REST API endpoints. REST Server library implements various component specific endpoints handling. Also, REST API Payload Parser is integrated using the “jsoncpp” OSS library to parse JSON format payload. New nvmultiurisrcbin implements callback APIs for runtime parameter configuration of the supported components in the native DeepStream gstreamer pipeline. Upon receiving the HTTP POST request for the REST API, relevant callbacks implemented with the nvmultiurisrcbin creates new custom gstreamer events which gets injected into the native gstreamer pipeline. Applicable gstreamer component intercepts the custom event to runtime configure the relevant component in the native gstreamer pipeline. A sample app deepstream-server-app is also provided at /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-server for reference. These interfaces sources are available in the DeepStream SDK package. This should provide users with readily available infrastructure to define and implement their own custom REST APIs support for various DeepStream components (Please refer “How to add custom REST API support” section below). REST server is instantiated by default inside nvmultiurisrcbin plugin.

DeepStream Rest Server block diagram

Refer to Gst-nvmultiurisrcbin (Alpha) for “nvmultiurisrcbin” details.

Features

The sample deepstream-server application can be used to demonstrate usage of REST APIs to configure DeepStream component parameters at runtime. Currently supported REST APIs to configure the DeepStream pipeline as mentioned below:

Note

DS-6.4 onwards version support has been introduced with REST API. Currently supported version with DS-6.4 is /api/v1.

DS-6.4 supports REST API error handling, as per OpenAPI Specification (https://spec.openapis.org/oas/v3.1.0).

The following table summarizes the supported REST APIs:

Rest API support

Group/Component name

Parameter update

Stream specific

Stream addition and removal

nvdspreprocess

Update ROI for preprocess

nvv4l2decoder

“drop-frame-interval” and “skip-frame” property update

nvdsinfer

“interval” property update

nvdsinferserver

“interval” property update

nvv4l2encoder

“force-idr”, “force-intra” and “bitrate” property update

nvstreammux

“batched-push-timeout” property update

nvvideoconvert

“src-crop”, “dest-crop”, “flip-method” and “interpolation-method” property update

nvdsosd

“process-mode” property update

Application specific

Application quit

nvds_rest_server

nvds_rest_server is a custom DeepStream library open sourced with DeepStreamSDK package at /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/. Users are encouraged to use this library as-is or make modifications to support custom use cases. This library could be integrated into a Gstreamer plugin, GStreamer bin, or a DeepStream application. With DS 6.2 release onwards, nvds_rest_server library is integrated with Gst-nvmultiurisrcbin (Alpha) to support runtime sensor addition and removal over REST API.

deepstream-server-app

The deepstream-server-app sample application is provided at /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-server. The reference application demonstrates REST API usage to configure DeepStream plugin parameters at runtime using nvmultiurisrcbin or using low level nvds_rest_server library APIs directly. Refer to the README at /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-server/README for application usage, payload schema and REST API endpoints details.

deepstream-server-app operates in two modes:

  1. REST server within nvmultiurisrcbin (default):

To use nvmultiurisrcbin, set within_multiurisrcbin: 1 in the rest-server group defined in dsserver_config.yml file. Paramters defined in multiurisrcbin group of dsserver_config.yml would be used by the application if simulation is done using nvmultiurisrcbin. All the REST API callbacks are implemented with nvmultiurisrcbin.

  1. REST server with the application:

To use REST server with the application, set within_multiurisrcbin: 0 in the dsserver_config.yml config file. Parameters defined inside group server-app-ctx of dsserver_config.yml would be used. All the REST API callbacks are implemented with the application.

Note

This application package contains dsserver_pgie_config.yml and dsserver_config.yml files. The dsserver_config.yml is used while launching the application. dsserver_pgie_config.yml is used to configure pgie in the sample application.

Expected behavior: The sample server app would run continuously even after EOS is received from the pipeline, as drop-pipeline-eos is set to TRUE (default ) in the config file dsserver_config.yml. However, if user needs to change this behavior, set ‘drop-pipeline-eos: 0’ of group ‘multiurisrcbin’ or ‘drop_pipeline_eos: 0’ of group ‘server-app-ctx’ defined in dsserver_config.yml file to exit app at EOS.

Sensor provisioning with deepstream-test5-app

The DeepStream config file parsing reference apps like deepstream-test5-app support Sensor provisioning (runtime stream add/remove). For more details and sample config file to refer, please follow documentation here.

How to add custom REST API support

Users should follow the below sections. Each section explains detailed steps to implement new custom REST API support.

  1. REST API endpoints implementation

  2. Custom event generation

  3. Callback implementation for REST API endpoints

REST API endpoints implementation

  1. Define any component/group specific custom REST API endpoints.

  2. Register custom REST API endpoints.

    For 1 & 2, refer below source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/nvds_rest_server.cpp:

if (uri.find ("/roi/update") != std::string::npos) {
  /* Pre-Process Specific */
  m_func[uri] =
      [roi_cb, uri] (const Json::Value & req_info, const Json::Value & in,
      Json::Value & out, struct mg_connection * conn) {
    return handleUpdateROI (req_info, in, out, conn, roi_cb, uri);
  };
}
  1. Define and implement the Handler function corresponding to the new custom REST API.

    Refer to the below source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/nvds_rest_server.cpp:

NvDsServerStatusCode
handleUpdateROI (const Json::Value & req_info, const Json::Value & in,
  Json::Value & response, struct mg_connection *conn,
  std::function < void (NvDsServerRoiInfo * roi_ctx, void *ctx) > roi_cb,
  std::string uri);

Refer to the Handler Function Implementation Details section for more details.

  1. Register the Handler function for the new custom REST API with Civerweb’s “httpServerHandler->addHandler” interface.

Refer to the below source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/nvds_rest_server.cpp:

httpServerHandler->addHandler (it.first, new RequestHandler (it.first,
        it.second));
  1. Define callback function to be used by REST server for the corresponding new custom REST API. The user needs to add callback function pointer as member of the struct NvDsServerCallbacks.

    Refer to the below source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/nvds_rest_server.h:

typedef struct NvDsServerCallbacks
{
  std::function < void (NvDsServerRoiInfo * roi_info, void *ctx) > roi_cb;
  std::function < void (NvDsServerDecInfo * dec_info, void *ctx) > dec_cb;
  std::function < void (NvDsServerEncInfo * enc_info, void *ctx) > enc_cb;
  std::function < void (NvDsServerStreamInfo * stream_info,
    void *ctx) > stream_cb;
  std::function < void (NvDsServerInferInfo * infer_info,
    void *ctx) > infer_cb;
  std::function < void (NvDsServerConvInfo * conv_info, void *ctx) > conv_cb;
  std::function < void (NvDsServerMuxInfo * mux_info, void *ctx) > mux_cb;
  std::function < void (NvDsServerInferServerInfo * inferserver_info,
    void *ctx) > inferserver_cb;
  std::function < void (NvDsServerOsdInfo * osd_info, void *ctx) > osd_cb;
  std::function < void (NvDsServerAppInstanceInfo * appinstance_info,
    void *ctx) > appinstance_cb;
  std::unordered_map <std::string, cb_func> custom_cb_endpt;
} NvDsServerCallbacks;

Handler Function Implementation Details

  1. Define and implement a parser function to parse payload data received with the HTTP request for the new custom REST API. User may define new custom data structure (with nvds_rest_server.h) to hold the parsed payload data.

    Refer to the source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/nvds_roi_parse.cpp

bool
nvds_rest_roi_parse (const Json::Value & in, NvDsServerRoiInfo * roi_info)
  1. Call the defined callback providing input the payload parsed data and custom context pointer (if any).

  2. Based on the callback function return status received, create a JSON response to be sent for the received HTTP request.

Refer to the below source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/nvds_rest_server/nvds_rest_server.cpp:

if (nvds_rest_roi_parse (in, &roi_info) && (roi_cb)) {
  roi_cb (&roi_info, &custom_ctx);
  switch (roi_info.roi_flag) {
    case ROI_UPDATE:
      http_err_code = NvDsServerStatusCodeToHttpStatusCode(roi_info.err_info.code);
      break;
    default:
      break;
  }
} else {
  http_err_code = NvDsServerStatusCodeToHttpStatusCode(roi_info.err_info.code);
}
res_info.status = std::string ("HTTP/1.1 ") + std::to_string (http_err_code.first) +
      " " + http_err_code.second;

res_info.reason = roi_info.roi_log;

response["status"] = res_info.status;
response["reason"] = res_info.reason;

Custom event generation

  1. Define and implement new custom gst-event applicable for specific “element or group of elements” corresponding to handling of the new custom REST API.

Refer to the below source code snippet in /opt/nvidia/deepstream/deepstream/sources/libs/gstnvdscustomhelper/gst-nvdscustomevent.c

GstEvent *
gst_nvevent_new_roi_update (gchar* stream_id, guint roi_count, RoiDimension *roi_dim)
  1. This custom gst-event to be injected into the native gstreamer pipeline by the gstnvdsmulturisrcbin.

Callback implementation for REST API endpoints

  1. Implement the callback function (as registered with the struct NvDsServerCallbacks) defined for the new custom REST API.

  1. If REST server is used with nvmultiurisrcbin, refer to the below source code snippet of /opt/nvidia/deepstream/deepstream/sources/gst-plugins/gst-nvmultiurisrcbin/gstdsnvmultiurisrcbin.cpp

static void s_roi_api_impl (NvDsServerRoiInfo * roi_info, void *ctx)
  1. If REST server is used with the application, refer to the below source code snippet of /opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-server/rest_server_callbacks.cpp

void s_roi_callback_impl (NvDsServerRoiInfo * roi_info, void *ctx)
  1. This callback function injects the new custom gst-event into the pipeline to be handled by corresponding gstreamer element or group of elements. Applicable gstreamer element or group of elements need to intercept this new custom gst-event at sink pad event handler and apply received configurations to the applicable element at runtime.

    Refer to the below source code snippet of /opt/nvidia/deepstream/deepstream/sources/gst-plugins/gst-nvmultiurisrcbin/gstdsnvmultiurisrcbin.cpp

GstEvent *nvevent = gst_nvevent_new_roi_update((char*)roi_info->stream_id.c_str(), roi_info->roi_count, roi_dim);
if (!nvevent) {
    roi_info->roi_log = "ROI_UPDATE_FAIL, nv-roi-update event creation failed";
    roi_info->status = ROI_UPDATE_FAIL;
    roi_info->err_info.code = StatusInternalServerError;
}

if (!gst_pad_push_event ((GstPad *) (nvmultiurisrcbin->bin_src_pad),
          nvevent)) {
  switch (roi_info->roi_flag) {
    case ROI_UPDATE:
      g_print ("[WARN] nv-roi-update event not pushed downstream.. !! \n");
      roi_info->roi_log = "ROI_UPDATE_FAIL, nv-roi-update event not pushed";
      roi_info->status = ROI_UPDATE_FAIL;
      roi_info->err_info.code = StatusInternalServerError;
      break;
    default:
      break;
  }
} else {
  switch (roi_info->roi_flag) {
    case ROI_UPDATE:
      roi_info->status = ROI_UPDATE_SUCCESS;
      roi_info->roi_log = "ROI_UPDATE_SUCCESS";
      roi_info->err_info.code = StatusOk;
      break;
    default:
      break;
  }
}
  1. In case the applicable element is already a part of the gstdsnvmultiurisrbin (e.g. decoder) then no such custom gst-events are required to be injected into the pipeline as all runtime configuration for the applicable gst-element can be handled from inside the gstdsnvmultiurisrcbin.

    Refer to the below source code snippet of /opt/nvidia/deepstream/deepstream/sources/gst-plugins/gst-nvmultiurisrcbin/gstdsnvmultiurisrcbin.cpp

if (!set_nvuribin_dec_prop (nvmultiurisrcbin->nvmultiurisrcbinCreator,
        sourceId, dec_info)) {
  switch (dec_info->dec_flag) {
    case DROP_FRAME_INTERVAL:
      g_print ("[WARN] drop-frame-interval not set on decoder .. !! \n");
      dec_info->status = DROP_FRAME_INTERVAL_UPDATE_FAIL;
      dec_info->dec_log = "DROP_FRAME_INTERVAL_UPDATE_FAIL, drop-frame-interval not set on decoder";
      dec_info->err_info.code = StatusInternalServerError;
      break;
    case SKIP_FRAMES:
      g_print ("[WARN] skip-frame not set on decoder .. !! \n");
      dec_info->status = SKIP_FRAMES_UPDATE_FAIL;
      dec_info->dec_log = "SKIP_FRAMES_UPDATE_FAIL, skip-frame not set on decoder";
      dec_info->err_info.code = StatusInternalServerError;
      break;
    case LOW_LATENCY_MODE:
      g_print ("[WARN] low-latency-mode not set on decoder .. !! \n");
      dec_info->status = LOW_LATENCY_MODE_UPDATE_FAIL;
      dec_info->dec_log = "LOW_LATENCY_MODE_UPDATE_FAIL, low-latency-mode not set on decoder";
      dec_info->err_info.code = StatusInternalServerError;
      break;
    default:
      break;
  }
} else {
  switch (dec_info->dec_flag) {
    case DROP_FRAME_INTERVAL:
      dec_info->status =
          dec_info->status !=
          DROP_FRAME_INTERVAL_UPDATE_FAIL ? DROP_FRAME_INTERVAL_UPDATE_SUCCESS
          : DROP_FRAME_INTERVAL_UPDATE_FAIL;
      if ( dec_info->status == DROP_FRAME_INTERVAL_UPDATE_SUCCESS ){
        dec_info->err_info.code = StatusOk;
        dec_info->dec_log = "DROP_FRAME_INTERVAL_UPDATE_SUCCESS";
      } else{
        dec_info->err_info.code = StatusInternalServerError;
        dec_info->dec_log = "DROP_FRAME_INTERVAL_UPDATE_FAIL, Error while setting drop-frame-interval property";
      }
      break;
    case SKIP_FRAMES:
      dec_info->status =
          dec_info->status !=
          SKIP_FRAMES_UPDATE_FAIL ? SKIP_FRAMES_UPDATE_SUCCESS :
          SKIP_FRAMES_UPDATE_FAIL;
      if ( dec_info->status == SKIP_FRAMES_UPDATE_SUCCESS ){
        dec_info->err_info.code = StatusOk;
        dec_info->dec_log = "SKIP_FRAMES_UPDATE_SUCCESS";
      } else{
        dec_info->err_info.code = StatusInternalServerError;
        dec_info->dec_log = "SKIP_FRAMES_UPDATE_FAIL, Error while setting skip-frame property";
      }
      break;
    case LOW_LATENCY_MODE:
      dec_info->status =
          dec_info->status !=
          LOW_LATENCY_MODE_UPDATE_FAIL ? LOW_LATENCY_MODE_UPDATE_SUCCESS :
          LOW_LATENCY_MODE_UPDATE_FAIL;
      if ( dec_info->status == LOW_LATENCY_MODE_UPDATE_SUCCESS ){
        dec_info->err_info.code = StatusOk;
        dec_info->dec_log = "LOW_LATENCY_MODE_UPDATE_SUCCESS";
      } else{
        dec_info->err_info.code = StatusInternalServerError;
        dec_info->dec_log = "LOW_LATENCY_MODE_UPDATE_FAIL, Error while setting skip-frame property";
      }
      break;
    default:
      break;
  }
}

Schema and endpoints

Note

DS-6.4 supports REST API version /api/v1. Refer below schema details.

Stream add/remove

  1. Stream add

Endpoint: /stream/add

Curl command to add stream:

1.curl -XPOST 'http://localhost:9000/api/v1/stream/add' -d '{
  "key": "sensor",
  "value": {
      "camera_id": "uniqueSensorID1",
      "camera_name": "front_door",
      "camera_url": "file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_1080p_h264.mp4",
      "change": "camera_add",
      "metadata": {
          "resolution": "1920 x1080",
          "codec": "h264",
          "framerate": 30
      }
  },
  "headers": {
      "source": "vst",
      "created_at": "2021-06-01T14:34:13.417Z"
  }
}'

2.curl -XPOST 'http://localhost:9000/api/v1/stream/add' -d '{
  "key": "sensor",
  "event": {
      "camera_id": "uniqueSensorID1",
      "camera_name": "front_door",
      "camera_url": "file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_1080p_h264.mp4",
      "change": "camera_streaming",
      "metadata": {
          "resolution": "1920 x1080",
          "codec": "h264",
          "framerate": 30
      }
  },
  "headers": {
      "source": "vst",
      "created_at": "2021-06-01T14:34:13.417Z"
  }
}'

Expected output: The uri specified should be added to the display. The camera_id should be unique for each newly added streams. The curl command mentioned above at (2), is only supported when the sample application uses nvmultiurisrcbin.

  1. Stream remove

Endpoint: /stream/remove

Curl command to remove stream

curl -XPOST 'http://localhost:9000/api/v1/stream/remove' -d '{
  "key": "sensor",
  "value": {
      "camera_id": "uniqueSensorID1",
      "camera_name": "front_door",
      "camera_url": "file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_1080p_h264.mp4",
      "change": "camera_remove",
      "metadata": {
          "resolution": "1920 x1080",
          "codec": "h264",
          "framerate": 30
      }
  },
  "headers": {
      "source": "vst",
      "created_at": "2021-06-01T14:34:13.417Z"
  }
}'

Expected output: The uri specified should be removed from the display. The camera_id used to remove stream should be same as being used while adding stream using REST API.

ROI

Endpoint: /roi/update

Curl command to update ROI:

curl -XPOST 'http://localhost:9000/api/v1/roi/update' -d '{
  "stream": {
      "stream_id": "0",
      "roi_count": 2,
      "roi": [{
              "roi_id": "0",
              "left": 100,
              "top": 300,
              "width": 400,
              "height": 400
          },
          {
              "roi_id": "1",
              "left": 550,
              "top": 300,
              "width": 500,
              "height": 500
          }
      ]
  }
}'

Expected output: The updated roi dimension should be observed at display.

Decoder

  1. Drop frame interval

Endpoint: /dec/drop-frame-interval

Configuration values for “drop_frame_interval” field of the schema: Range [0 - 30] Curl command to configure decoder drop-frame-interval property:

curl -XPOST 'http://localhost:9000/api/v1/dec/drop-frame-interval' -d '{
"stream":
  {
      "stream_id":"0",
      "drop_frame_interval":2
  }
}'

Expected output: The drop-frame-interval value will be set on the decoder. Decoder drop frame interval should reflect with every interval <value> frame given by decoder, rest all dropped for selected stream.

  1. Skip frame

Endpoint: /dec/skip-frames

Configuration values for “skip_frames” field of the schema:

(0): - Decode all frames (1): - Decode non-ref frames (2): - Decode key frames

Curl command to configure decoder skip-frames property

curl -XPOST 'http://localhost:9000/api/v1/dec/skip-frames' -d '{
"stream":
  {
      "stream_id":"0",
      "skip_frames":2
  }
}'

Expected output: The skip-frames property value will be set on the decoder.

(0): - Decoder will decode all frames of the encoded bitstream (1): - Decoder will decode only non-reference frames of the encoded bitstream (2): - Decoder will decode only key frames of the encoded bitstream

Nvinfer

Endpoint: /infer/set-interval

Curl command to configure nvinfer interval property

curl -XPOST 'http://localhost:9000/api/v1/infer/set-interval' -d '{
"stream":
  {
      "stream_id":"0",
      "interval":2
  }
}'

Expected output: The interval value will be set on the nvinfer. Interval value specify consecutive batches will be skipped for inference for the video stream.

Disable/comment “input-tensor-meta” property in dsserver_pgie_config.yml to see “interval” property functionality of nvinfer/nvinferserver. Currently stream_id (specified in the schema) do not have any impact on specified stream_id, rather configuration is getting applied to all active streams.

Nvinferserver

Endpoint:/inferserver/set-interval

Curl command to configure nvinferserver interval property:

curl -XPOST 'http://localhost:9000/api/v1/inferserver/set-interval' -d '{
"stream":
  {
      "stream_id":"0",
      "interval":2
  }
}'

Expected output: The interval value will be set on nvinferserver. Interval value specify consecutive batches will be skipped for inference for the video stream.

Currently stream_id (specified in the schema) do not have any impact on specified stream_id, rather configuration is getting applied to all active streams.

Encoder

Note

By default encoder is disabled. To enable, set enable: 1 in the “encoder” group of dsserver_config.yml. Currently stream_id (specified in the schema) do not have any impact on specified stream_id, rather configuration is applied on muxed encoded bitstream.

  1. Force-idr

Endpoint: /enc/force-idr

Configuration value for “force_idr” field of the schema:

(1): - Force IDR frame

Curl command to configure encoder force idr frame property:

curl -XPOST 'http://localhost:9000/api/v1/enc/force-idr' -d '{
"stream":
  {
      "stream_id":"0",
      "force_idr":1
  }
}'

Expected output: The force-idr property value will be set on the encoder. Encoder force-idr property should reflect with insertion of the IDR frame with the encoded bitstream by the encoder.

  1. Force-intra

Endpoint: /enc/force-intra

Configuration value for “force_intra” field of the schema:

(1): - Force Intra frame

Curl command to configure encoder force intra frame property

curl -XPOST 'http://localhost:9000/api/v1/enc/force-intra' -d '{
"stream":
  {
      "stream_id":"0",
      "force_intra":1
  }
}'

Expected output: The force-intra property value will be set on the encoder. Encoder force-intra property should reflect with insertion of the intra frame with the encoded bitstream by the encoder.

  1. Bitrate

Endpoint: /enc/bitrate

Curl command to configure encoder bitrate property

curl -XPOST 'http://localhost:9000/api/v1/enc/bitrate' -d '{
"stream":
  {
      "stream_id":"0",
      "bitrate":2000000
  }
}'

Convert generated .h264 elementary bitstream to mp4 file using below commands:

$ ffmpeg -i out.h264 -vcodec copy out.mp4
$ mediainfo out.mp4

Expected output: Encoder should be reconfigured to use updated bitrate <value> and provide corresponding encoded bitstream. Mediainfo should show Encoder bitrate corresponding to updated value.

Streammux

Endpoint: /mux/batched-push-timeout

Configuration value for “batched_push_timeout” field of the schema:

(microseconds): - Timeout value

Curl command to configure streammux batched pushed timeout property:

curl -XPOST 'http://localhost:9000/api/v1/mux/batched-push-timeout' -d '{
"stream":
  {
      "batched_push_timeout":100000
  }
}'

Applicable for old nvstreammux. Expected output: The batched push timeout property value will be set on the nvstreammux. nvstreammux property should reflect with the timeout in microseconds to wait after the first buffer is available to push the batch even if the complete batch is not formed.

Nvvideoconvert

Note

To simulate video convert specific REST API features, deepstream-server application explicitly disables passthrough mode using the “disbale-passthrough” property of nvvideoconvert within the nvmultiurisrcbin. Set disable-passthrough: 1 in dsserver_config.yml file.

  1. src-crop

Endpoint: /conv/srccrop Configuration value for “src_crop” field of the schema: (String) Pixel location left:top:width:height

Curl command to configure nvvideoconvert src-crop property:

curl -XPOST 'http://localhost:9000/api/v1/conv/srccrop' -d '{
"stream":
  {
      "stream_id":"0",
      "src_crop":"200:200:400:500"
  }
}'

Expected output: left:top:width:height of the input image which will be cropped and transformed into the output buffer. If the crop location is out of bound the values will be clamped to image boundaries of the input image.

  1. dest-crop

Endpoint: /conv/destcrop

Configuration value for “dest_crop” field of the schema: (String) Pixel location left:top:width:height

Curl command to configure nvvideoconvert dest-crop property:

  curl -XPOST 'http://localhost:9000/api/v1/conv/destcrop' -d '{
  "stream":
    {
        "stream_id":"0",
        "dest_crop":"100:200:400:500"
    }
  }'

Expected output: left:top:width:height is the location in the output image
where the input image will be transformed.  If the crop location is out of
bound the values will be clamped to image boundaries of the output image.
The region apart from the cropped location in the destination frame will
retain the last pixel values.
  1. flip-method

Endpoint: /conv/flip-method

Configuration value for “flip_method” field of the schema:

(0): none             - Identity (no rotation)
(1): counterclockwise - Rotate counter-clockwise 90 degrees
(2): rotate-180       - Rotate 180 degrees
(3): clockwise        - Rotate clockwise 90 degrees
(4): horizontal-flip  - Flip horizontally
(5): upper-right-diagonal - Flip across upper right/lower left diagonal
(6): vertical-flip    - Flip vertically
(7): upper-left-diagonal - Flip across upper left/lower right diagonal

Curl command to configure nvvideoconvert flip-method property:

curl -XPOST 'http://localhost:9000/api/v1/conv/flip-method' -d '{
"stream":
  {
      "stream_id":"0",
      "flip_method":2
  }
}'

Expected output: Based on flip-method property type value, output image should be flipped. For ex- For value 2, image will be rotated by 180 degree.

  1. interpolation-method

Endpoint: /conv/interpolation-method

Configuration value for “interpolation_method” field of the schema:

(0): Nearest          - Nearest
(1): Bilinear         - Bilinear
(2): Algo-1           - GPU - Cubic, VIC - 5 Tap
(3): Algo-2           - GPU - Super, VIC - 10 Tap
(4): Algo-3           - GPU - LanzoS, VIC - Smart
(5): Algo-4           - GPU - Ignored, VIC - Nicest
(6): Default          - GPU - Nearest, VIC - Nearest

Curl command to configure nvvideoconvert interpolation-method property:

curl -XPOST 'http://localhost:9000/api/v1/conv/interpolation-method' -d '{
"stream":
  {
      "stream_id":"0",
      "interpolation_method":2
  }
}'

Expected output: There would not be any visual change, but applied interpolation-method should be used for transformation.

Nvdsosd

Endpoint: /osd/process-mode

Configuration value for “process_mode” field of the schema:

0 and 1, 0=CPU mode, 1=GPU mode

Curl command to configure nvdsosd process_mode property:

curl -XPOST 'http://localhost:9000/api/v1/osd/process-mode' -d '{
"stream":
  {
      "stream_id":"0",
      "process_mode":0
  }
}'

Expected output: There would not be any visual change, but applied process-mode should be used for drawing bounding boxes.

Application Instance

Application quit

Endpoint: /app/quit

Configuration value for “app_quit” field of the schema: (1): - Application quit (boolean)

Curl command to quit the sample application:

curl -XPOST 'http://localhost:9000/api/v1/app/quit' -d '{
"stream":
  {
      "app_quit":1
  }
}'

Expected output: The application should quit.

Limitations

  1. REST API feature is validated with HTTP. HTTPS support is not yet enabled, however users can add HTTPS support by enhancing corresponding sources in DeepStream SDK.

  2. New nvstreammux support for REST API is currently in alpha mode.

  3. REST API support for nvvidconvert and nvstreammux is currently only supported in mode where nvmultiurisrcbin is used in the pipeline. Refer dsserver_config.yml file present at deepstream-server-app source directory.

  4. Flip-method property of nvvideoconvert is supported with legacy nvstreammux only.

  5. MJPEG input stream source is not supported.

  6. Decoder REST API for skip-frame is not supported on Jetson. Currently on x86, skip-frame for value “(1): decode_non_ref” is not supported.