main menu

• Members Main Page
• Search for AMVs
• Forum
• How-To Guides
• Site FAQ
• About us
• Login

Information

• Member: Haunter103
• Title: The Gullable Mermaid
• Premiered: 2025-05-23
• Categories:
  • Comedy
  • Character Profile
  • Fun
• Song:
  • Bad Lip Reading Seagulls! (Stop It Now)
• Anime:
  • Muromi-san (TV)
  • Namiuchigiwa no Muromi-san (OVA)
• Participation:
  • Anime North, Anime North AMV Contest (2025-05-24)
• Comments: So, this one is over 8 years in the making. I have premiere project file auto-saves dating back to January of 2017. The original YouTube video the audio is sourced from premiered in November of 2016. So it's been brewing for a while.
  
  I wasn't super concerned someone would make a video with this song while I dragged my heels on it, since I figured the anime I used wasn't well known and probably the only thing that would work, and I knew I'd have to bend it a lot to get it to work. The anime came out in 2013. Its main character is 'Muromi-san', a mermaid who is constantly pestering the fishing boy 'Takkun'. They were perfect stand-ins for the versions of Yoda and Luke depicted in the Bad Lip Reading - Seagulls video. One is a person who's a little unstable and the other is a mildly irate straight-man. Also I think it goes a long way that Muromi happens to have green hair to connect her to yoda just that much more.
  
  The concept sat for so long due primarily to a general lack of motivation on my part to do any editing in my spare time, and the difficulty of the scene selection process - having to dip back into the well of episodes repeatedly for clips that came at all close to meeting a series of very specific criteria. I was a week into a 2-week vacation I'd scheduled in 2019 before I had the drive to crack open premiere to make any real progress after starting things off two years prior, and it was years in between me taking additional runs at filling out the timeline.
  
  But it was always in the back of my mind and the seagulls song kept playing in my head and occasionally through my car's audio system through via Bluetooth from my phone (I bought the song on iTunes as soon as I started the project shortly after seeing the Bad lip reading vid for the first time, back when iTunes was somewhat relevant). And I managed to finally settle in all the clips I wanted to use on the sequence in the fall of 2024. At that point the difficult creative decisions that were triggering my procrastination instincts for so many years were out of the way, and I knew it was just the hard work of all the lipsync and grueling compositing work that was keeping me from the finish line. Since I always want to submit to my home con, Anime North, that set for me a deadline of about 6-7 months in late March of 2025. I deemed this to be achievable. I was most right.. although I had to work in another 1-week vacation the week of the entry deadline.
  
  Submitted to AN; some friends recommended I also submit to Momocon, so I did, and I made some improvements and additions to the effects that I didn't have time for on the AN deadline. I made most improvements/tweaks/additions after that as well, because I figured if it was going to take me 8 years to finish it, I might as well make it as perfect as It feasibly can to minimize regrets before I put it up online - the point of no return.
  
  
  For more detail on the effects... it was a lot.
  
  Regarding lipsync, I did my usual pattern of just editing a scene the way I think works best with as much dynamic movement in the characters as I can get away with. I don't shy away from those kind of shots and enjoy the challenge lipsyncing them gives me. I had a lot of places where I needed to draw custom mouths though, primarily for the "Mmmm" sounds Yoda/Muromi makes throughout. I know I couldn't just get away with regular mouths on those, they needed that comic punch, but I'm not much of an artist, and it took a minute before I found the right reference for what they should look like in a frame of Akane from the first episode of the Ranma 1/2 remake anime.
  
  For compositing, there was a boatload of shots where I chose to remove elements from the frame that I deemed to be distracting. Characters and objects that served a gag in the context of the original anime, but sort of confused the matter or broke up the flow/disturbed the continuity for what I was going for. So I got rid of them with a combination of Content-Aware/Generative Fill functions in Photoshop, as well as masking and other compositing methods in After Effects. There was a fair amount of background replacement as well, since I wanted to keep things consistent for portions of the video that were intended to be taking place in the same space. I didn't want the dock to be shown if the video is depicting a flashback of Muromi getting attacked on the beach for instance. I also went as far as to alter Takkun's shirt colour in multiple shots just to ease continuity in those same-space sections. Couldn't do anything about his shirt type changing constantly though unfortunately.
  
  Then there were a few cases where I had to go even further and create elements from scratch. I wanted these to look like they were native to the series if possible. I didn't really want them standing out in the way foreign objects tend to when you simply import a jpg and keyframe it. That has a different kind of jokey feel to it than what I was aiming to accomplish. The hacky sack is an example, and I created that from scratch (google image search for visual reference) using shape layers in AE that I keyframed over the existing footage manually frame-by-frame to animate it squishing into Muromi's cheek. But there was also a shot where I needed a little girl to point and laugh. However, in the original shot, she was completely static, and her arm wasn't out (and she even lacked a mouth). So I had to break her apart in after effects and puppet-tool warp her into motion, while drawing a new arm extending to her shoulders with shape layers (traced an entirely different frame of an arm of Muromi pointing to something), and keyframe motion on that. And drew some new mouth frames from nothing all to get a short laughing animation cycle that didn't exist. Pretty proud of that one.
  
  One more scene I want to spotlight is (actually two scenes) where Muromi in shown on Takkun's back while he runs. One of these involved a background replacement and both of them required either a full or partial rotoscope of the boy in order to have Muromi appear behind him. But due to the fact that they were moving from the background to the foreground in both cases, I realized that simply scaling up the footage would disrupt the effective compositing of Muromi, since her line thickness would also scale up, and suddenly be much thinker than Takkun's by the end of the shot. The only way for me to account for this was to vector-trace all 19 frames of Muromi's head/hair in the shot I used for both scenes (which is taken from the clip I used for "..groove and boogie." - since her hair is swaying back and forth so I knew it would fit the other scenes perfectly as she bobs around on his back). SO that took a while but I knew that work wouldn't be doable before either AMV contest submission deadline, so I backburnered it until later for online release. But the result was what I was aiming for in the end.
  
  I should also mention that AI was used in the making of this video. The Fill tools in Photoshop are a mild example of this, but there were a couple of instances where I created something from scratch in Photoshop using prompts. These were - the rock that one of the seagulls is perched on in the first scene we see Muromi getting attacked by seagulls, and the crater in the concrete of the pier/dock seen after the 'boulder' crashes down (although the compositing and distortion effects add to that crater were all me). But I thought it was worth mentioning.
  Also I think this video might not have been possible for me to complete much earlier than I did regardless, because there was a small handful of key scenes that were Standard Definition quality-only and required AI-upscaling software to be at all usable. These were from the OVA of the series, which was only ever available on DVD. Part of why it took so long was I was long expecting I'd have to replace all these OVA scene's I'd temped in on the timeline eventually, but there was never anything that made for suitable substitutes. So while I'm not a proponent of Artificial Intelligence in the broader scope of creative applications, I do have to concede there are some AI tools that can be useful, and I daresay this AMV doesn't turn out the way it did without them.
  
  
  I managed to come away with the Northern Lights Award for Best Overall Video at Anime North, and was voted Best Comedy at Momocon, also receiving the "Phade's Phav" award. I was touched that the founder of this site and the person who made the first AMV I ever saw which helped get me into the hobby liked what I'd made. So I guess I can say taking a full week break from my job in video post production only to spend the entirety of that time still deep within Adobe softwares paid off.
  
  
  Partially dedicated to my cousin. The seagulls thing was a family in-joke that resulted from a game of Taboo. She's very competitive; this was years before the song or even the anime existed.