diff --git a/libavcodec/mpegvideo_enc.c b/libavcodec/mpegvideo_enc.c
index 9dd30e6a2d..44c93953c3 100644
--- a/libavcodec/mpegvideo_enc.c
+++ b/libavcodec/mpegvideo_enc.c
@@ -962,8 +962,9 @@ static int load_input_picture(MpegEncContext *s, const AVFrame *pic_arg)
     Picture *pic = NULL;
     int64_t pts;
     int i, display_picture_number = 0, ret;
-    const int encoding_delay = s->max_b_frames ? s->max_b_frames :
-                                                 (s->low_delay ? 0 : 1);
+    int encoding_delay = s->max_b_frames ? s->max_b_frames
+                                         : (s->low_delay ? 0 : 1);
+    int flush_offset = 1;
     int direct = 1;
 
     if (pic_arg) {
@@ -1075,11 +1076,22 @@ static int load_input_picture(MpegEncContext *s, const AVFrame *pic_arg)
 
         pic->f->display_picture_number = display_picture_number;
         pic->f->pts = pts; // we set this here to avoid modifiying pic_arg
+    } else {
+        /* Flushing: When we have not received enough input frames,
+         * ensure s->input_picture[0] contains the first picture */
+        for (flush_offset = 0; flush_offset < encoding_delay + 1; flush_offset++)
+            if (s->input_picture[flush_offset])
+                break;
+
+        if (flush_offset <= 1)
+            flush_offset = 1;
+        else
+            encoding_delay = encoding_delay - flush_offset + 1;
     }
 
     /* shift buffer entries */
-    for (i = 1; i < MAX_PICTURE_COUNT /*s->encoding_delay + 1*/; i++)
-        s->input_picture[i - 1] = s->input_picture[i];
+    for (i = flush_offset; i < MAX_PICTURE_COUNT /*s->encoding_delay + 1*/; i++)
+        s->input_picture[i - flush_offset] = s->input_picture[i];
 
     s->input_picture[encoding_delay] = (Picture*) pic;