Scribd
Upload
32 views10 pages

Khưu Kiên An - 1751048: Group Project 1: Fashion Smart Recommendation (Group 08)

This personal report summarizes three group projects: 1. A fashion recommendation project that used a dataset of outfits to build a bidirectional LSTM model to predict compatible outfits. It comments that compatibility ratings could consider fashion experts' opinions. 2. A food image to recipe project that aimed to provide recipes for edible foods by comparing image and recipe vectors using cosine similarity loss. It suggests adding more recipes. 3. A WiFi-based fall detection project that sought a generalized deep learning model, but provides no details on its dataset or methods. The report has no comments on this project.

Uploaded by

Kien An
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
32 views10 pages

Khưu Kiên An - 1751048: Group Project 1: Fashion Smart Recommendation (Group 08)

This personal report summarizes three group projects: 1. A fashion recommendation project that used a dataset of outfits to build a bidirectional LSTM model to predict compatible outfits. It comments that compatibility ratings could consider fashion experts' opinions. 2. A food image to recipe project that aimed to provide recipes for edible foods by comparing image and recipe vectors using cosine similarity loss. It suggests adding more recipes. 3. A WiFi-based fall detection project that sought a generalized deep learning model, but provides no details on its dataset or methods. The report has no comments on this project.

Uploaded by

Kien An
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

WR227 - Personal report

Khưu Kiên An – 1751048

Group project 1: Fashion smart recommendation (group 08)

+Objective:

Authors’s goal is not to waste too much time, select the best outfit for clients,
make the world interesting.

+Dataset:

They collect dataset from Xintong Han from Polyvore containing 21,889 outfits.
For outfits having too many fashion items, it’s kept the first 8 for simplicity. The
resulting Polyvore dataset contains 164,379 items (each item contains product
image and a text description).

The average of fashion item in an outfit is 6.5. And fashion items in an outfit on
Polyvore.com organized in fixed order: tops, bottoms, shoes and accessories.

Link: https://github.com/xthan/polyvore-dataset
+Method:

They using bidirectional LSTM for fashion compatibility modeling and a visual-
semantic embedding to capture multimodal information.

+Personal comment:

Fashion clothes somehow are hot to wear, so I think they might add more
problem like clothes suit for weather, and about clothes set rating (compatiblity),
they take the average fashion point of fashionistas, fashion magazine, or just
themselves’s opinion.

Group project 2: Food image to cooking recipe (group 10)

+Objective:

Authors’s objective is to give cooking recipes for each edible food.

+Dataset:
+Method:

They compare 2 vectors (Two vectors are for compute the cosine similarity loss
which will rate how fit is their collected recipe and image) from image and recipe,
then edit and choose the fittest recipe. And they use cosine similarity loss to test
and train data; LSTM; image, ingredients, instructions encoder
+Personal comment:

Maybe food recipe can be limited, so I think they can add more recipes to
continue the project

Group project 3: Wifi fall detection (group 03)

+Objective:

Their objective is looking for a generalized deep learning model for Wi-Fi based
fall detection in untrained domains.

+Dataset:
+Method:
+Personal comment:

Their prestation is well-explained. I don’t have anything to discuss them.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy