… To maintain the advantages of using synthetic data while avoiding its negative effects, we
propose to introduce geometry-aware contrastive learning to learn multiview consistent …

… Regarding computational efficiency, this is where contrastive learning has an advantage
over our proposed NeRF-based decoder, as the encoding with CURL can be trained within half …

… based on semantic contrast learning, which takes Neural Radiance Fields (NeRF) as the 3D
… by leveraging the semantic capabilities of the Contrastive Language-Image Pre-Training (…

… Following prior works [27, 57], we leverage contrastive learning methods to train a ResNet-50
[21] as visual encoder. The visual and tactile encoders share the same architecture but …

… Then, we introduce our global-local contrastive learning … a contrastive learning strategy to
control the stylization strength (Fig. 3(c)). Specifically, in the framework of contrastive learning, …

… and learn the aforementioned crucial 3D spatial field consistency, in this paper, we propose
a Contrastive In-Voxel Transformer (CVT) structure to implement the 3D field consistency in …

… this contrastive approach because it allows for defining pairwise relationships between points
based on mask labels. However, we design a scale-conditioned contrastive … field can learn …

… To render a CLIP embedding along a ray, the field is … scale field which allows hierarchical
text queries. Grad-CAM [33], attention-based methods [7], or patchaligned contrastive learning […

… a contrastive learning-based pretraining strategy to improve the feature extractor. (2) We reduce
the geometric complexity of the object by learning … to learn a generalizable radiance field …

… neural radiance fields (NeRF) with a convolutional encoder to learn … fields in parallel to the
RGB radiance fields in NeRF to learn … representation for reinforcement learning. SNeRL outper…