Merge pull request #14 from tsaishien-chen/main

Fix captioning inference and more_sample page bugs

captioning/inference.py

@@ -23,6 +23,13 @@ class DotDict(dict):
     parser.add_argument("--video-list", required=True, help="list of input videos.")
     parser.add_argument("--output-json", default=None, help="output json file. Leave none to print out the results.")
     parser.add_argument("--prompt-list", default=None, help="list of correponding input prompts. Leave none if no prompt input.")
+    parser.add_argument(
+        "--options",
+        nargs="+",
+        help="override some settings in the used config, the key-value pair "
+        "in xxx=yyy format will be merged into config file (deprecate), "
+        "change to --cfg-options instead.",
+    )
     args = parser.parse_args()
     cfg = Config(args)
 

docs/more_samples.html

@@ -303,7 +303,7 @@ <h2 class="pt-4"><p class="text-center" id="Food">Food</p></h2>
 	<p style="font-family: Chalkduster; font-size: 16px; color: white">"A person mixing chocolate in a bowl with an electric mixer."</p>
 </th>
 <th style="text-align: center; vertical-align: top; padding: 10px;">
-	<img src="./assets/samples/_ONhUIKHX_4.4.gif" style="width: 100%"></img>
+	<video playsinline autoplay loop muted src="./assets/samples/ONhUIKHX_4.4.mp4" style="width: 100%" type="video/mp4"></video>
 	<p style="font-family: Chalkduster; font-size: 16px; color: white">"There are sausages cooking on a grill, and a person is using tongs to turn them over."</p>
 </th>
 </tr>
@@ -564,7 +564,7 @@ <h2 class="pt-4"><p class="text-center" id="Tutorial_and_Narrative">Tutorial and
 	<p style="font-family: Chalkduster; font-size: 16px; color: white">"There are several glasses filled with different types of drinks on a counter."</p>
 </th>
 <th style="text-align: center; vertical-align: top; padding: 10px;">
-	<img src="./assets/samples/_uQs-YDb5VA.9.gif" style="width: 100%"></img>
+	<video playsinline autoplay loop muted src="./assets/samples/uQs-YDb5VA.9.mp4" style="width: 100%" type="video/mp4"></video>
 	<p style="font-family: Chalkduster; font-size: 16px; color: white">"A person in blue gloves is connecting an electrical supply to an injector."</p>
 </th>
 </tr>