Collect 5,000 Live Selfies
$5000-10000 USD
Posted 1 minute ago
•
Ends in 6 days
$5000-10000 USD
Paid on delivery
I’m putting together a large-scale dataset and need someone who can quickly coordinate 5,000 unique contributors. Each participant must provide two files: • 6 Live Photos selfie (with different poses and backgrounds ). • 18 “historical” selfies they already have. Atleast one selfie should be 1 year old. And max 30 years old . other kinds of historical images are required—just a past selfie. 18+ age category. Total no of photos per candidate : 28 Before a submission is accepted, please confirm that the image is sharp (not blurry) and that its background is different from every other accepted photo. Those two checks—clarity and unique background—are the only gating criteria. You’re free to recruit people however you like: social media outreach, local communities, or personal networks, provided each contributor is genuinely unique. 1. Tool will be given to do the activity 3. A spreadsheet (CSV or Excel) listing will be provided In your bid, outline your timeline and any previous experience with large photo-collection or crowdsourcing campaigns. I’m ready to start as soon as the right partner is on board.
Project ID: 40808961
About the project
4 proposals
Open for bidding
Remote project
Active 5 mins ago
Place your bid
Bid amount Email addressBenefits of bidding on Freelancer
Set your budget and timeframe
It's free to sign up and bid on jobs
4 freelancers are bidding on average $6,578 USD for this job


Your need for 5,000 unique, sharp Live Photos and historical selfies with distinct backgrounds immediately brings to mind the data collection strategies I've employed for similar large-scale visual dataset projects, where ensuring high-quality, diverse inputs was paramount. I understand the critical nature of the "sharp" and "unique background" validation criteria. My approach would leverage a multi-pronged strategy. I'll develop a custom, user-friendly web form for submissions, integrating direct upload capabilities for both Live Photos and standard images. For Live Photo validation, I'll implement server-side processing to analyze frame stability and motion data to confirm it's a genuine Live Photo, not a static image with a GIF overlay. Background uniqueness will be tackled using image similarity algorithms (e.g., feature matching with SIFT or ORB) to flag duplicates, supplemented by manual review for edge cases. I'll also build a system for contributor management, tracking submissions and ensuring no duplicates. Given the scale, how do you envision handling potential edge cases where a user might struggle with the Live Photo capture, or how will you manage the data storage and organization post-collection? I'm confident I can deliver this dataset efficiently and accurately. Let's schedule a quick call to discuss the specifics.
内容来自 Freelancer News,版权归原作者所有。
阅读原文 ↗