ڧgL 6 d dl Z d dlZd dlmZmZ d dlmZ d dlmZ d dlm Z m
Z
m Z m Z m
Z
d dlZd dlZd dlmZ ddlmZmZ d d
lmZmZ d d lmZ e ej ej e ej e ej f Z e ej ej e ej f Z d Z G d
dee Z G d de Z G d de Z! G d de Z" G d de Z# G d de Z$de%dej fdZ&de%de ej ej f fd Z'dS ) N)ABCabstractmethod)globPath)CallableListOptionalTupleUnion)Image )
decode_png read_file ) _read_pfmverify_str_arg)
VisionDataset) KittiFlowSintelFlyingThings3D FlyingChairsHD1Kc e Zd ZdZddeeef dee ddf fd
Z dede
j
fdZ e defd Z
d
edeeef fd Zdefd Zd
edej j j fdZ xZS ) FlowDatasetFNroot
transformsreturnc v t | || _ g | _ g | _ d S )N)r )super__init__r
_flow_list _image_list)selfr r __class__s ^/var/www/html/ai-engine/env/lib/python3.11/site-packages/torchvision/datasets/_optical_flow.pyr! zFlowDataset.__init__# s:
d###$%',. file_namec n t j | }|j dk r| d }|S )NRGB)r
openmodeconvert)r$ r( imgs r& _read_imgzFlowDataset._read_img+ s4 j## 8u ++e$$C
r' c d S N r$ r( s r&
_read_flowzFlowDataset._read_flow1 s
r' indexc | | j | d }| | j | d }| j r0| | j | }| j r|\ }}nd }nd x}}| j | |||| \ }}}}| j s|||||fS |||fS )Nr r )r/ r# r" r4 _has_builtin_flow_maskr )r$ r5 img1img2flowvalid_flow_masks r& __getitem__zFlowDataset.__getitem__6 s ~~d .u 5a 899~~d .u 5a 899 ? *??4?5#9::D*
'(,%oo"&%) )D? ? &04dDRa0b0b -D$o & $/*Et_4 4t# #r' c * t | j S r1 )lenr# )r$ s r& __len__zFlowDataset.__len__M s 4#$$$r' vc R t j j | g|z S r1 )torchutilsdata
ConcatDataset)r$ r@ s r& __rmul__zFlowDataset.__rmul__P s! { --tfqj999r' r1 )__name__
__module__ __qualname__r7 r strr r
r r! r
r/ r r4 intT1T2r< r? rB rC rD rE rF
__classcell__r% s @r& r r s( #/ /U39 - /8H;M /Y] / / / / / /3 5;
C
^
$ $r2v $ $ $ $.% % % % %:# :%+"2"@ : : : : : : : :r' r c e Zd ZdZ ddeeef dededee d df
fd
Z d e
d ee e f f fd Z
d
ed ej fdZ xZS )r a `Sintel `_ Dataset for optical flow.
The dataset is expected to have the following structure: ::
root
Sintel
testing
clean
scene_1
scene_2
...
final
scene_1
scene_2
...
training
clean
scene_1
scene_2
...
final
scene_1
scene_2
...
flow
scene_1
scene_2
...
Args:
root (str or ``pathlib.Path``): Root directory of the Sintel Dataset.
split (string, optional): The dataset split, either "train" (default) or "test"
pass_name (string, optional): The pass to use, either "clean" (default), "final", or "both". See link above for
details on the different passes.
transforms (callable, optional): A function/transform that takes in
``img1, img2, flow, valid_flow_mask`` and returns a transformed version.
``valid_flow_mask`` is expected for consistency with other datasets which
return a built-in valid mask, such as :class:`~torchvision.datasets.KittiFlow`.
traincleanNr split pass_namer r c t || t |dd t |dd |dk rdd gn|g}t | d
z }|d z d z }|D ]}|d
k rd n|}||z |z }t j | D ]} t
t t || z dz }
t t |
dz
D ]$} | xj |
| |
| dz ggz
c_ %|d
k r=| xj t
t t || z dz z
c_ d S )Nr r rS rQ test valid_valuesrT rR finalbothr] rR r\ r trainingr: rQ *.pngr *.flo)
r r! r r oslistdirsortedr rJ ranger> r# r" )
r$ r rS rT r passes flow_root split_dir
image_rootscene
image_listir% s
r& r! zSintel.__init__} s dz:::ug4EFFFFy+`__ dataset for optical flow (2015).
The dataset is expected to have the following structure: ::
root
KittiFlow
testing
image_2
training
image_2
flow_occ
Args:
root (str or ``pathlib.Path``): Root directory of the KittiFlow Dataset.
split (string, optional): The dataset split, either "train" (default) or "test"
transforms (callable, optional): A function/transform that takes in
``img1, img2, flow, valid_flow_mask`` and returns a transformed version.
TrQ Nr rS r r c N t || t |dd t | dz |dz z }t t t
|dz dz }t t t
|dz d z }|r|st d
t || D ]\ }}| xj ||ggz
c_ |d k r6t t t
|d z dz | _
d S d S )
NrV rS rW rY r ingimage_2z*_10.pngz*_11.pngzZCould not find the Kitti flow images. Please make sure the directory structure is correct.rQ flow_occ) r r! r r rc r rJ FileNotFoundErrorzipr# r" ) r$ r rS r images1images2r8 r9 r% s r& r! zKittiFlow.__init__ s=
dz:::ug4EFFFFDzzK'55=9c$"2Z"? @ @AABBc$"2Z"? @ @AABB g #l
gw// / /JD$ $ . G $T#dZ.?*.L*M*M%N%N O ODOOO r' r5 c F t | S )a Return example at given index.
Args:
index(int): The index of the example to retrieve
Returns:
tuple: A 4-tuple with ``(img1, img2, flow, valid_flow_mask)``
where ``valid_flow_mask`` is a numpy boolean mask of shape (H, W)
indicating which flow values are valid. The flow is a numpy array of
shape (2, H, W) and the images are PIL images. ``flow`` and ``valid_flow_mask`` are None if
``split="test"``.
rn ro s r& r< zKittiFlow.__getitem__ rp r' r( c t | S r1 )_read_16bits_png_with_flow_and_valid_maskr3 s r& r4 zKittiFlow._read_flow 8CCCr' rQ N)rG rH rI rv r7 r rJ r r
r r! rK rL rM r< r rw rx r4 rN rO s @r& r r s & "P PU39 - Pc PQYZbQc Pos P P P P P P(
*
*r2v
*
*
*
*
*
* DC DE"*bj2H,I D D D D D D D Dr' r c e Zd ZdZd
deeef dedee ddf fd
Z d e
dee e f f fd
Z
d edej fd Z xZS )r a `FlyingChairs `_ Dataset for optical flow.
You will also need to download the FlyingChairs_train_val.txt file from the dataset page.
The dataset is expected to have the following structure: ::
root
FlyingChairs
data
00001_flow.flo
00001_img1.ppm
00001_img2.ppm
...
FlyingChairs_train_val.txt
Args:
root (str or ``pathlib.Path``): Root directory of the FlyingChairs Dataset.
split (string, optional): The dataset split, either "train" (default) or "val"
transforms (callable, optional): A function/transform that takes in
``img1, img2, flow, valid_flow_mask`` and returns a transformed version.
``valid_flow_mask`` is expected for consistency with other datasets which
return a built-in valid mask, such as :class:`~torchvision.datasets.KittiFlow`.
rQ Nr rS r r c t || t |dd t | dz }t t t
|dz dz }t t t
|dz dz }d }t j ||z st d
t j t
||z t j
}t t | D ]a}|| } |d k r| d
k s |dk rE| dk r?| xj || gz
c_ | xj |d|z |d|z d
z ggz
c_ bd S )NrV rS )rQ valrY r rD z*.ppmr` zFlyingChairs_train_val.txtzmThe FlyingChairs_train_val.txt file was not found - please download it from the dataset page (see docstring).)dtyperQ r r r )r r! r r rc r rJ ra pathexistsr~ rw loadtxtint32rd r> r" r# ) r$ r rS r imagesflowssplit_file_name
split_listrk split_idr% s r& r! zFlyingChairs.__init__ s
dz:::ug4DEEEEDzzN*S !8 9 9::;;tCv
7 8 899::6w~~d_ 455 #
ZD?$: ; ;28LLL
s5zz"" I IA!!}H X]]8WX==E!H:- fQUmVAEAI5F%G$HH I Ir' r5 c F t | S )a Return example at given index.
Args:
index(int): The index of the example to retrieve
Returns:
tuple: A 3-tuple with ``(img1, img2, flow)``.
The flow is a numpy array of shape (2, H, W) and the images are PIL images.
``flow`` is None if ``split="val"``.
If a valid flow mask is generated within the ``transforms`` parameter,
a 4-tuple with ``(img1, img2, flow, valid_flow_mask)`` is returned.
rn ro s r& r< zFlyingChairs.__getitem__ rp r' r( c t | S r1 rr r3 s r& r4 zFlyingChairs._read_flow( rt r' r ru rO s @r& r r s 2I IU39 - Ic IQYZbQc Ios I I I I I I.
*
*r2v
*
*
*
*
*
* $C $BJ $ $ $ $ $ $ $ $r' r c
e Zd ZdZ ddeeef deded ed
ee d df fd
Z d
e
d ee e f f fd Z
ded ej fdZ xZS )r a `FlyingThings3D `_ dataset for optical flow.
The dataset is expected to have the following structure: ::
root
FlyingThings3D
frames_cleanpass
TEST
TRAIN
frames_finalpass
TEST
TRAIN
optical_flow
TEST
TRAIN
Args:
root (str or ``pathlib.Path``): Root directory of the intel FlyingThings3D Dataset.
split (string, optional): The dataset split, either "train" (default) or "test"
pass_name (string, optional): The pass to use, either "clean" (default) or "final" or "both". See link above for
details on the different passes.
camera (string, optional): Which camera to return images from. Can be either "left" (default) or "right" or "both".
transforms (callable, optional): A function/transform that takes in
``img1, img2, flow, valid_flow_mask`` and returns a transformed version.
``valid_flow_mask`` is expected for consistency with other datasets which
return a built-in valid mask, such as :class:`~torchvision.datasets.KittiFlow`.
rQ rR leftNr rS rT camerar r c
t || t |dd | }t |dd dgdgddgd| }t d d
d k rd d
gng}t | dz }d}t j ||| D ]\ }t t t ||z |z dz } t fd| D } t t t |dz |z dz }
t fd|
D }
| r|
st d t | |
D ]\ } } t t t | dz }
t t t | dz }t t | dz
D ]}dk r:| xj |
| |
|dz ggz
c_ | xj || gz
c_ Bdk r<| xj |
|dz |
| ggz
c_ | xj ||dz gz
c_ d S )NrV rS rW rY rT r[ frames_cleanpassframes_finalpassr )r rightr] r] r r r ) into_future into_pastz*/*c 3 < K | ]}t | z V d S r1 r ).0 image_dirr s r& z*FlyingThings3D.__init__..e s. U UYY& 8 U U U U U Ur' optical_flowc 3 B K | ]}t | z z V d S r1 r )r flow_dirr directions r& r z*FlyingThings3D.__init__..h s3 ] ]xtH~~ 9F B ] ] ] ] ] ]r' zcCould not find the FlyingThings3D flow images. Please make sure the directory structure is correct.r_ z*.pfmr r r )r r! r upperr itertoolsproductrc r rJ r~ r rd r> r# r" )r$ r rS rT r r re cameras
directions
image_dirs flow_dirsr r r r rk r r% s ` @r& r! zFlyingThings3D.__init__I s dz:::ug4EFFFF
y+ >?? tC7(:$;$; < <== s5zzA~.. : :A M11 ((fQiA -G,HH(( E!H:5"k11 ((fQUmVAY-G,HH(( E!a%L>9
:
: : :r' r5 c F t | S rm rn ro s r& r< zFlyingThings3D.__getitem__{ rp r' r( c t | S r1 )r r3 s r& r4 zFlyingThings3D._read_flow rt r' )rQ rR r Nru rO s @r& r r , s > )-
0: 0:CI 0: 0: 0:
0: X&
0:
0: 0: 0: 0: 0: 0:d
*
*r2v
*
*
*
*
*
* $C $BJ $ $ $ $ $ $ $ $r' r c e Zd ZdZdZddeeef dedee ddf fd
Z
d
ede e j
e j
f fd Zd edeeef f fd
Z xZS )r a `HD1K `__ dataset for optical flow.
The dataset is expected to have the following structure: ::
root
hd1k
hd1k_challenge
image_2
hd1k_flow_gt
flow_occ
hd1k_input
image_2
Args:
root (str or ``pathlib.Path``): Root directory of the HD1K Dataset.
split (string, optional): The dataset split, either "train" (default) or "test"
transforms (callable, optional): A function/transform that takes in
``img1, img2, flow, valid_flow_mask`` and returns a transformed version.
TrQ Nr rS r r c
t || t |dd t | dz }|dk rt d D ]}t t
t |dz d z |d
d z }t t
t |d z d
z |d
d z }t t | dz
D ];}| xj || gz
c_ | xj
|| ||dz ggz
c_
r" r# r r~ )
r$ r rS r seq_idxr r rk r r image1image2r% s
r& r! z
HD1K.__init__ s
dz:::ug4EFFFFDzzF" G 99
E
E tC~(=
(JPWMcMcMcMc(c$d$d e eff S )r s # # # # # # # # 9 9 9 9 9 9 9 9 9 9 9 9 9 9 , , , , , , , , , , , , , , , , ! ! ! ! ! !
5; Xbj%98BJ;O OP
5; Xbj%9 9: 4: 4: 4: 4: 4:#} 4: 4: 4:nT$ T$ T$ T$ T$[ T$ T$ T$n:D :D :D :D :D :D :D :DzA$ A$ A$ A$ A$; A$ A$ A$H_$ _$ _$ _$ _$[ _$ _$ _$D@* @* @* @* @*; @* @* @*F8 8 8 8 8 8"1 1rzSUS]G]A^ 1 1 1 1 1 1r'