h-dZddlmZddlmZmZddlZddlZddlmZddl m Z ddl m Z dd l mZdd lmZmZmZmZdd lmZdd lmZd dlmZej4eZeedGddeZeGdde ZddZGddej@Z!Gddej@Z"edGddeZ#ddgZ$y)zPyTorch VitPose model.) dataclass)OptionalUnionN)nn)BackboneOutput)PreTrainedModel)Unpack) ModelOutputTransformersKwargsauto_docstringlogging) load_backbone)can_return_tuple) VitPoseConfigz6 Class for outputs of pose estimation models. ) custom_introceZdZUdZdZeejed<dZ eejed<dZ ee ejdfed<dZ ee ejdfed<y)VitPoseEstimatorOutputaH loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided): Loss is not supported at this moment. See https://github.com/ViTAE-Transformer/ViTPose/tree/main/mmpose/models/losses for further detail. heatmaps (`torch.FloatTensor` of shape `(batch_size, num_keypoints, height, width)`): Heatmaps as predicted by the model. hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`): Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each stage) of shape `(batch_size, sequence_length, hidden_size)`. Hidden-states (also called feature maps) of the model at the output of each stage. Nlossheatmaps. hidden_states attentions) __name__ __module__ __qualname____doc__rrtorch FloatTensor__annotations__rrtuplerj/var/www/html/eduruby.in/venv/lib/python3.12/site-packages/transformers/models/vitpose/modeling_vitpose.pyrr&sq )-D(5$$ %,,0Hhu(()0=AM8E%"3"3S"89:A:>Ju00#567>r#rcxeZdZUeed<dZdZdZdee je je jffdZ y)VitPosePreTrainedModelconfigvit pixel_valuesTmodulect|tjtjfrtjj |j jjtjd|jjj|j j|j _|j%|jjjyyt|tj rJ|jjj|j jj#dyy)zInitialize the weightsg)meanstdNg?) isinstancerLinearConv2dinit trunc_normal_weightdatatorfloat32r'initializer_rangedtypebiaszero_ LayerNormfill_)selfr*s r$ _init_weightsz$VitPosePreTrainedModel._init_weightsEs fryy"))4 5"$!6!6 ""%%emm43DKKDaDa"7"b$$% MM {{&   &&('  - KK   " " $ MM   $ $S ).r#N)rrrrr base_model_prefixmain_input_namesupports_gradient_checkpointingrrr/r0r;r>r"r#r$r&r&>s= $O&*# *E"))RYY *L$M *r#r&c|dvr td|jdk7r td|j\}}}}d}|dk(rd}|dddddd f |dddddd f<|j|d |||}|j }|j D])\} } |dd| d f|dd| d f<|dd| d f|dd| d f<+|j||||f}|j d }|S) aFlip the flipped heatmaps back to the original form. Args: output_flipped (`torch.tensor` of shape `(batch_size, num_keypoints, height, width)`): The output heatmaps obtained from the flipped images. flip_pairs (`torch.Tensor` of shape `(num_keypoints, 2)`): Pairs of keypoints which are mirrored (for example, left ear -- right ear). target_type (`str`, *optional*, defaults to `"gaussian-heatmap"`): Target type to use. Can be gaussian-heatmap or combined-target. gaussian-heatmap: Classification target with gaussian distribution. combined-target: The combination of classification target (response map) and regression target (offset map). Paper ref: Huang et al. The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation (CVPR 2020). Returns: torch.Tensor: heatmaps that flipped back to the original image )gaussian-heatmapcombined-targetz9target_type should be gaussian-heatmap or combined-targetzCoutput_flipped should be [batch_size, num_keypoints, height, width]rrDrN.) ValueErrorndimshapereshapeclonetolistflip) output_flipped flip_pairs target_type batch_size num_keypointsheightwidthchannelsoutput_flipped_backleftrights r$ flip_backrYTs8"AATUUa^__/=/C/C,J vuH''(6q!$Q$|(D'Dq!$Q$|$#++JHfeTN(..0"((*J e,:1eS=,IAtSL)-;AtSL-IAucM*J.55z=RXZ_6`a-2226 r#ceZdZdZdeffd ZddejdeejdejfdZ xZ S) VitPoseSimpleDecoderz Simple decoding head consisting of a ReLU activation, 4x upsampling and a 3x3 convolution, turning the feature maps into heatmaps. r'ct|tj|_tj |j dd|_tj|jj|jddd|_ y)NbilinearF) scale_factormode align_cornersrr kernel_sizestridepadding) super__init__rReLU activationUpsampler^ upsamplingr0backbone_config hidden_size num_labelsconvr=r' __class__s r$rfzVitPoseSimpleDecoder.__init__se '')++63F3FZglmII  " " . .0A0AqYZde  r# hidden_staterOreturnc|j|}|j|}|j|}| t||}|SN)rhrjrnrYr=rqrOrs r$forwardzVitPoseSimpleDecoder.forwardsC|4 |4 99\*  ! :6Hr#rt rrrrrrfrTensorrrv __classcell__rps@r$r[r[|sB  }  ELL hu||>T `e`l`l r#r[cheZdZdZdeffd ZddejdeejfdZ xZ S)VitPoseClassicDecoderz Classic decoding head consisting of a 2 deconvolutional blocks, followed by a 1x1 convolution layer, turning the feature maps into heatmaps. r'ct|tj|jj ddddd|_tjd|_tj|_ tjdddddd|_ tjd|_ tj|_ tjd|jddd|_y) NrErF)rbrcrdr9rra)rerfrConvTranspose2drkrldeconv1 BatchNorm2d batchnorm1rgrelu1deconv2 batchnorm2relu2r0rmrnros r$rfzVitPoseClassicDecoder.__init__s ))  " " . .1VW^c  ..-WWY ))#s!UV]bc ..-WWY IIc6#4#4!AWXY r#rqrOc|j|}|j|}|j|}|j|}|j |}|j |}|j |}| t||}|Srt)rrrrrrrnrYrus r$rvzVitPoseClassicDecoder.forwards{||L1 |4 zz,/ ||L1 |4 zz,/ 99\*  ! :6Hr#rtrwrzs@r$r|r|s6 Z} ZELLhu||>Tr#r|z? The VitPose model with a pose estimation head on top. ceZdZdeffd Zee d dejde ejde ejde ejde e de f d Z xZS) VitPoseForPoseEstimationr'ct||t||_t |jj ds t dt |jj ds t dt |jj ds t d|jr t|n t||_ |jy)Nrlz0The backbone should have a hidden_size attribute image_sizez0The backbone should have an image_size attribute patch_sizez/The backbone should have a patch_size attribute) rerfrbackbonehasattrr'rGuse_simple_decoderr[r|head post_initros r$rfz!VitPoseForPoseEstimation.__init__s  %f- t}}++];OP Pt}}++\:OP Pt}}++\:NO O4:4M4M(0ShioSp  r#r) dataset_indexrOlabelskwargsrrc d}| td|jj|fd|i|}|jd}|jd} |j j jd|j j jdz} |j j jd|j j jdz} |jddd}|j| d| | j}|j||} t|| |j|j S) ac dataset_index (`torch.Tensor` of shape `(batch_size,)`): Index to use in the Mixture-of-Experts (MoE) blocks of the backbone. This corresponds to the dataset index used during training, e.g. For the single dataset index 0 refers to the corresponding dataset. For the multiple datasets index 0 refers to dataset A (e.g. MPII) and index 1 refers to dataset B (e.g. CrowdPose). flip_pairs (`torch.tensor`, *optional*): Whether to mirror pairs of keypoints (for example, left ear -- right ear). Examples: ```python >>> from transformers import AutoImageProcessor, VitPoseForPoseEstimation >>> import torch >>> from PIL import Image >>> import requests >>> processor = AutoImageProcessor.from_pretrained("usyd-community/vitpose-base-simple") >>> model = VitPoseForPoseEstimation.from_pretrained("usyd-community/vitpose-base-simple") >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg" >>> image = Image.open(requests.get(url, stream=True).raw) >>> boxes = [[[412.8, 157.61, 53.05, 138.01], [384.43, 172.21, 15.12, 35.74]]] >>> inputs = processor(image, boxes=boxes, return_tensors="pt") >>> with torch.no_grad(): ... outputs = model(**inputs) >>> heatmaps = outputs.heatmaps ```NzTraining is not yet supportedrrFrrr)rO)rrrr)NotImplementedErrorrforward_with_filtered_kwargs feature_mapsrIr'rkrrpermuterJ contiguousrrrr) r=r)rrOrrroutputssequence_outputrQ patch_height patch_widthrs r$rvz VitPoseForPoseEstimation.forwards>N  %&EF F"L$--"L"L # '# # "..r2$**1- {{22==a@DKKD_D_DjDjklDmm kk11< ll>   -> U\\* > & > +, >  > > r#r)rC)%r dataclassesrtypingrrrtorch.utils.checkpointrmodeling_outputsrmodeling_utilsr processing_utilsr utilsr r r rutils.backbone_utilsr utils.genericrconfiguration_vitposer get_loggerrloggerrr&rYModuler[r|r__all__r"r#r$rs!" .-&MM1-0   H %   ?[?  ?$*_***%P2996#BII#L S 5S  S l $%? @r#