From 4297a6750593a3b29c88e296343a74283c2315e3 Mon Sep 17 00:00:00 2001 From: haruishi <7902640+haruishi43@users.noreply.github.com> Date: Sun, 7 Apr 2024 11:43:56 +0900 Subject: [PATCH] Update README.md --- README.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/README.md b/README.md index ea407552..8a09e7fc 100644 --- a/README.md +++ b/README.md @@ -20,6 +20,7 @@ - Compatible with `cuda` tensors for faster processing. - No need for other dependencies except for `numpy` and `torch`. - Added functionality like creating rotation matrices, batched processing, and automatic type detection. +- Works with various input modals - Highly modular If you found this module helpful to your project, please site this repository: @@ -73,6 +74,11 @@ There are no _real_ differences in `class` or `func` APIs: Each API automatically detects the input type (`numpy.ndarray` or `torch.Tensor`), and outputs are the same type. +The arguments for each `class` or `func` depends on the transform, but here are the common arguments: +- `z_down (bool)`: whether to use a coordinate system with z-axis pointing down, defaults to `False` +- `mode (str)`: interpolation mode, defaults to `bilinear` +- `clip_output (bool)`: whether to clip values based on the range of the input values, default to `True` + An example for `Equi2Pers`/`equi2pers`: