U-hrnet: delving into improving semantic representation of high resolution network for dense prediction
High resolution and advanced semantic representation are both vital for dense prediction.
Empirically, low-resolution feature maps often achieve stronger semantic representation …
Empirically, low-resolution feature maps often achieve stronger semantic representation …
Densely connected multi-dilated convolutional networks for dense prediction tasks
N Takahashi, Y Mitsufuji - … of the IEEE/CVF conference on …, 2021 - openaccess.thecvf.com
Tasks that involve high-resolution dense prediction require a modeling of both local and
global patterns in a large input field. Although the local and global structures often depend …
global patterns in a large input field. Although the local and global structures often depend …
Efficientvit: Lightweight multi-scale attention for high-resolution dense prediction
High-resolution dense prediction enables many appealing real-world applications, such as
computational photography, autonomous driving, etc. However, the vast computational cost …
computational photography, autonomous driving, etc. However, the vast computational cost …
Fapn: Feature-aligned pyramid network for dense image prediction
Recent advancements in deep neural networks have made remarkable leap-forwards in
dense image prediction. However, the issue of feature alignment remains as neglected by …
dense image prediction. However, the issue of feature alignment remains as neglected by …
Generalizing interactive backpropagating refinement for dense prediction networks
As deep neural networks become the state-of-the-art approach in the field of computer vision
for dense prediction tasks, many methods have been developed for automatic estimation of …
for dense prediction tasks, many methods have been developed for automatic estimation of …
Rethinking local and global feature representation for dense prediction
Although fully convolution networks (FCNs) have dominated dense prediction tasks (eg,
semantic segmentation, depth estimation and object detection) for decades, they are …
semantic segmentation, depth estimation and object detection) for decades, they are …
Polymax: General dense prediction with mask transformer
Dense prediction tasks, such as semantic segmentation, depth estimation, and surface
normal prediction, can be easily formulated as per-pixel classification (discrete outputs) or …
normal prediction, can be easily formulated as per-pixel classification (discrete outputs) or …
SSA: Semantic structure aware inference for weakly pixel-wise dense predictions without cost
Y Sun, Z Li - arXiv preprint arXiv:2111.03392, 2021 - arxiv.org
The pixel-wise dense prediction tasks based on weakly supervisions currently use Class
Attention Maps (CAM) to generate pseudo masks as ground-truth. However, the existing …
Attention Maps (CAM) to generate pseudo masks as ground-truth. However, the existing …
Multi-task learning with multi-query transformer for dense prediction
Previous multi-task dense prediction studies developed complex pipelines such as multi-
modal distillations in multiple stages or searching for task relational contexts for each task …
modal distillations in multiple stages or searching for task relational contexts for each task …
EfficientViT: Lightweight multi-scale attention for on-device semantic segmentation
High-resolution dense prediction enables many appealing real-world applications, such as
computational photography, autonomous driving, etc. However, the vast computational cost …
computational photography, autonomous driving, etc. However, the vast computational cost …