mlsanityData to 7 Oct 2026

Multiple View Performers for Shape Completion

arXiv 2209.06291 ↗First posted 13 Sep 2022cs.CV cs.ROcited by 0

Not on the map. Why?

Abstract (arXiv)

We propose the Multiple View Performer (MVP) - a new architecture for 3D shape completion from a series of temporally sequential views. MVP accomplishes this task by using linear-attention Transformers called Performers. Our model allows the current observation of the scene to attend to the previous ones for more accurate infilling. The history of past observations is compressed via the compact associative memory approximating modern continuous Hopfield memory, but crucially of size independent from the history length. We compare our model with several baselines for shape completion over time, demonstrating the generalization gains that MVP provides. To the best of our knowledge, MVP is the first multiple view voxel reconstruction method that does not require registration of multiple depth views and the first causal Transformer based model for 3D shape completion.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 53 references that are in the map. First posted on arXiv 13 Sep 2022.

Builds on

29 of 53 references.

References · 29

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  4. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  5. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  6. 2020Rethinking Attention with Performerscited by 1,584
  7. 20163D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstructioncited by 587
  8. 2017Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classificationcited by 521
  9. 2020Hopfield Networks is All You Needcited by 461
  10. 2019DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusioncited by 352
  11. 2022CLIP-Mesh: Generating textured meshes from text using pretrained image-text modelscited by 281
  12. 2019What Do Single-view 3D Reconstruction Networks Learn?cited by 259
17 more
  1. 2016Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesiscited by 219
  2. 2019Interactive Gibson Benchmark (iGibson 0.5): A Benchmark for Interactive Navigation in Cluttered Environmentscited by 89
  3. 2016Shape Completion Enabled Robotic Graspingcited by 79
  4. 2022CenterSnap: Single-Shot Multi-Object 3D Shape Reconstruction and Categorical 6D Pose and Size Estimationcited by 69
  5. 2015Gated Feedback Recurrent Neural Networkscited by 54
  6. 2021Multi-view 3D Reconstruction with Transformercited by 52
  7. 2018Robust Attentional Aggregation of Deep Feature Sets for Multi-view 3D Reconstructioncited by 44
  8. 2021MVT: Multi-view Vision Transformer for 3D Object Recognitioncited by 34
  9. 2016A note on the triangle inequality for the Jaccard distancecited by 32
  10. 2018Dense 3D Object Reconstruction from a Single Depth Viewcited by 24
  11. 2019Render4Completion: Synthesizing Multi-View Depth Maps for 3D Shape Completioncited by 22
  12. 2022Chefs' Random Tables: Non-Trigonometric Random Featurescited by 20
  13. 2020Sub-Linear Memory: How to Make Performers SLiMcited by 18
  14. 2021MC-LSTM: Mass-Conserving LSTMcited by 17
  15. 2018Photometric Depth Super-Resolutioncited by 8
  16. 2021Mobile Manipulation Leveraging Multiple Viewscited by 4
  17. 2020Attention-based 3D Object Reconstruction from a Single Imagecited by 2

Built on by

Not cited yet.