MCPcopy Create free account

hub / github.com/WU-CVGL/Omni-Scene / types & classes

Types & classes33 in github.com/WU-CVGL/Omni-Scene

⨍ Functions 190◇ Types & classes 33↳ Endpoints 1
↓ 5 callersClassNetLinLayer
A single linear layer which does a 1x1 conv
model/losses.py:95
↓ 4 callersClassAttention
model/volume/vit.py:172
↓ 3 callersClassKVCompressCrossViewAttnProcessor
model/pixel/attention_processor.py:131
↓ 2 callersClassLPIPS
model/losses.py:38
↓ 1 callersClassBenchmarker
model/utils/benchmarker.py:11
↓ 1 callersClassBlock
model/volume/vit.py:254
↓ 1 callersClassDepth2Normal
model/gaussian.py:84
↓ 1 callersClassGaussianRenderer
model/gaussian.py:117
↓ 1 callersClassMlp
MLP as used in Vision Transformer, MLP-Mixer and related networks.
model/volume/vit.py:223
↓ 1 callersClassPatchEmbed
Image to Patch Embedding.
model/volume/vit.py:311
↓ 1 callersClassScalingLayer
model/losses.py:85
↓ 1 callersClassvgg16
model/losses.py:104
ClassDummyLoss
model/losses.py:158
ClassKVCompressAttnProcessor
model/pixel/attention_processor.py:19
ClassLN2d
A LayerNorm variant, popularized by Transformers, that performs pointwise mean and variance normalization over the channel dimension for input
model/volume/vit.py:17
ClassLossDepthTV
model/losses.py:12
ClassMVDownsample2D
model/pixel/blocks.py:39
ClassMVMiddle2D
model/pixel/blocks.py:235
ClassMVUpsample2D
model/pixel/blocks.py:133
ClassOmniGaussian
model/omni_gs.py:25
ClassPixelGaussian
model/pixel/pixel_gs.py:22
ClassSpecifyGradient
model/utils/ops.py:55
ClassTPVCrossViewHybridAttention
TPVFormer Cross-view Hybrid Attention Module.
model/volume/cross_view_hybrid_attention.py:16
ClassTPVFormerEncoder
model/volume/tpvformer_encoder.py:14
ClassTPVFormerLayer
Base `TPVFormerLayer` for vision transformer. It can be built from `mmcv.ConfigDict` and support more flexible customization, for example, us
model/volume/tpvformer_layer.py:14
ClassTPVFormerPositionalEncoding
model/volume/positional_encoding.py:9
ClassTPVImageCrossAttention
An attention module used in TPVFormer. Args: embed_dims (int): The embedding dimension of Attention. Default: 256. nu
model/volume/image_cross_attention.py:15
ClassTPVMSDeformableAttention3D
An attention module used in tpvFormer based on Deformable-Detr. `Deformable DETR: Deformable Transformers for End-to-End Object Detection. <h
model/volume/image_cross_attention.py:188
ClassViT
Vision Transformer with support for patch or hybrid CNN input stage.
model/volume/vit.py:345
ClassVolumeGaussian
model/volume/volume_gs.py:12
ClassVolumeGaussianDecoder
model/volume/volume_gs_decoder.py:11
Class_TruncExp
model/utils/ops.py:39
ClassnuScenesDataset
data/dataloader.py:79