-
### 🚀 The feature
Addition of new / Fine Tuning of existing models to support OCR for `Handwritten Text`.
As a first step we can start with detection/prediction models that work specifically for Han…
-
Hi,
I tried out the algorithm to test on a few streetview images and am getting shorter heights than expected.
Learning from the earlier issues on Github, I used the customized 'vps_models/config…
-
I'm creating a real-time pitch detector with the help of CREPE.
If the model is learned on 16 000 Hz data and takes as input 1024 samples, is it true that the maximum real-time updating speed of t…
-
## 🚀 Feature
Uniform representation of masks in predictions and annotations
## Motivation
As it became apparent, for training it is required to add mask annotations using `target.add_field("mas…
botcs updated
5 years ago
-
I find that there are some bad cases in F0 prediction. I recommend people to increase 'predictor_layers' or decrease 'predictor_dropout' to enhance the ability of pitch predictor for the part of MIDI …
-
### 🚀 The feature, motivation and pitch
Currently, we can use `to_captum` to explain one prediction example only. However, it would be good to be able to explain predictions in batches as well.
Cur…
-
# Task Name
Japanese Pitch Accent Recognition
## Task Objective
Pitch accent is an important feature in the Japanese language that distinguishes between words through variations in pitch. Eac…
-
## 🚀 Feature
YOLOv5 uses 3 fixed transforms during Test Time Augmentations (TTA): 3 scales with a left-right flip on the middle scale. Our users are requesting specific transforms for TTA (i.e. rot…
-
### 🚀 The feature
Add the AVA dataset ([AVA: A Large-Scale Database for Aesthetic Visual Analysis](http://refbase.cvc.uab.es/files/MMP2012a.pdf)) in TorchVision.
### Motivation, pitch
It curr…
-
## 🚀 Feature
Add NMS version which takes confidence threshold and prunes out low confidence scores from the entire prediction.
## Motivation
Motivation is that if Pytorch NMS requires only boxes …