Update to use new run script for Intel GPUs.

docker-compose.nodeodm.gpu.intel.yml

@@ -10,7 +10,9 @@ services:
     environment:
       - WO_DEFAULT_NODES
   node-odm:
-    image: opendronemap/nodeodm:gpu.intel.local
+    environment:
+      - RENDER_GROUP_ID
+    image: opendronemap/nodeodm:gpu.intel
     devices:
       - "/dev/dri"
     ports:

webodm.sh

@@ -54,17 +54,16 @@ detect_gpus(){
 	fi
 }
 
-prepare_intel_gpu_image(){
+prepare_intel_render_group(){
 	if [ "${GPU_INTEL}" = true ]; then
-		curl -O https://raw.githubusercontent.com/OpenDroneMap/NodeODM/master/Dockerfile.gpu.intel.local
-		RENDER_GROUP_ID=$(getent group render | cut -d":" -f3)
-		docker build -f Dockerfile.gpu.intel.local -t opendronemap/nodeodm:gpu.intel.local --build-arg RENDER_GROUP_ID="${RENDER_GROUP_ID}" .
-		rm Dockerfile.gpu.intel
+		export RENDER_GROUP_ID=$(getent group render | cut -d":" -f3)
+	else
+		export RENDER_GROUP_ID=0
 	fi
 }
 
 detect_gpus
-prepare_intel_gpu_image
+prepare_intel_render_group
 
 case $key in
     --port)