[fix] minor bug - for image-based remote inference pipeline

Author: chyomin06

compressai_vision/pipelines/remote_inference/image_remote_inference.py

@@ -148,8 +148,8 @@ def __call__(
             timing["decode"].append((end - start))
 
             start = time_measure()
-            # dec_d = {"file_name": dec_seq["file_names"][0]}
-            dec_d = {"file_name": dec_seq[0]["file_names"][0]}
+            dec_d = {"file_name": dec_seq["file_names"][0]}
+            # dec_d = {"file_name": dec_seq[0]["file_names"][0]}
 
             pred = vision_model.forward(org_map_func(dec_d))
             end = time_measure()