Method and system for predicting garment attributes using deep learning

System for predicting garment attributes using deep learning techniques, the system including a processor configured to:

• (i) receive and store one or more digital image datasets including images of training garments;
• (ii) train a deep model for garment attribute identification, using the stored one or more digital image datasets, by configuring a deep neural network model to predict
• (a) multiple-class discrete attributes;
• (b) binary discrete attributes, and
• (c) continuous attributes,
• (iii) receive one or more digital images of a garment, wherein the one or more digital images of the garment are not present in the one or more digital image datasets, and
• (iv) extract attributes of the garment from the one or more received digital images of the garment using the trained deep model for garment attribute identification, wherein the extracted attributes include one or more of:
• multiple-class discrete attributes, binary discrete attributes, continuous attributes;
• wherein the processor is configured to:
• receive user information, including one or more of:
• user's body shape parameters, user's location, age, and ethnicity;
• receive garment sizing and measurement information, including one or more of:
• garment sizes, size-charts of the garment, garment measurements;
• receive fit-style labels relating to the training garment images in the one or more digital image datasets, including one or more of:
• the circumferential fits over different body parts and vertical drops, and
• re-architect and fine-tune the trained deep model.

A computer-implemented method of digitising a garment, and estimating the physics parameters of fabric material of the garment, the method using a garment digitization apparatus, the apparatus including a mannequin, a mannequin rotation system, a computer system and a camera system, the method including the steps of:

• (i) imaging the mannequin wearing the garment using the camera system;
• (ii) rotating the mannequin wearing the garment through at least 360° using the mannequin rotation system;
• (iii) capturing at least three images of the garment using the camera system during the mannequin rotation,
• (iv) generating fast and jerky left-right-left rotations at a series of configured rotational accelerations and velocities to disturb the garment on the mannequin with patterned motion, and
• (v) capturing garment appearance under the motion and estimating the physics parameters of the garment fabric material.

A system for digitising a garment, and estimating physics parameters of fabric material of the garment, the system including a garment digitization apparatus, the apparatus including a mannequin, a mannequin rotation system, a computer system and a camera system, the system arranged to:

• (i) image the mannequin wearing the garment using the camera system;
• (ii) rotate the mannequin wearing the garment through at least 360° using the mannequin rotation system;
• (iii) capture at least three images of the garment using the camera system during the mannequin rotation,
• (iv) generate fast and jerky left-right-left rotations at a series of configured rotational accelerations and velocities to disturb the garment on the mannequin with patterned motion, and
• (v) capture a garment appearance under the motion and estimate the physics parameters of the garment fabric material.

Computer implemented method for predicting garment attributes using deep learning techniques, comprising the steps of:

• (i) receiving and storing one or more digital image datasets including images of training garments;
• (ii) training a deep model for garment attribute identification, using the stored one or more digital image datasets, by configuring a deep neural network model to predict
• (a) multiple-class discrete attributes;
• (b) binary discrete attributes, and
• (c) continuous attributes,
• (iii) receiving one or more digital images of a garment, wherein the one or more digital images of the garment are not present in the one or more digital image datasets, and
• (iv) extracting attributes of the garment from the one or more received digital images of the garment using the trained deep model for garment attribute identification, wherein the extracted attributes include one or more of:
• multiple-class discrete attributes, binary discrete attributes, continuous attributes;
• and further including the steps of:
• (I) collecting one or more real photos and one or more synthetic rendered images,
• (II) training the deep model to generate a difference image,
• (III) using the deep model to generate a difference image, and
• (IV) superposing the difference image onto an input synthetic rendered image to generate a photo-realistic synthetic image.