Drexubery/ViewCrafter

Issues

• Reproducing Video Generation from RealEstate10K as in Fig. 3

  #29 opened 2 months ago by sixiaozheng
  2

• Quantitative evaluation of Sparse NVS task

  #32 opened 2 months ago by zhanghaoyu816
  2

• Reproducing the quantitative results on RealEstate10K

  #33 opened 2 months ago by XiangZ-0
  3

• How to do iterative optimization of point cloud?

  #43 opened 2 months ago by zero-kappa
  1

• About GPU Memories reported in README.md

  #23 opened 2 months ago by Yukun-Huang
  11

• Reference images used for conditioning

  #42 opened 2 months ago by Sisso16
  2

• What is the base model of Viewcrafter and it's training strategy?

  #40 opened 2 months ago by liziye12348765
  2

• Confused about the video pairs organization

  #38 opened 2 months ago by everythoughthelps
  3

• multi gpu setting

  #39 opened 2 months ago by sp77-1
  1

• dynamic video NVS

  #37 opened 2 months ago by zhw-zhang
  4

• CUDA out of memory with Default Script

  #36 opened 2 months ago by hbb1
  2

• The problem of ckpt_25_512

  #35 opened 2 months ago by Kidleyh
  2

• Questions about training

  #30 opened 2 months ago by xiyichen
  4

• Color Inconsistency with 4 sparse views

  #34 opened 2 months ago by LaFeuilleMorte
  2

• Sparse view

  #31 opened 2 months ago by itsss
  1

• training code

  #27 opened 2 months ago by miaolin968
  3

• Question about reconstruction strategy in iterative novel view synthesis

  #21 opened 2 months ago by tsuJohc
  5

• About training data process

  #28 opened 2 months ago by cyf23
  1

• using camera 4*4 matrix from dust3r instead of theta, phi, r

  #16 opened 3 months ago by saetlan
  4

• question about 360° rendering.

  #19 opened 3 months ago by Vincento-Wang
  3

• Some question about FrozenOpenCLIPImageEmbedderV2

  #10 opened 3 months ago by Kidleyh
  9

• Questions about training

  #1 opened 4 months ago by xizaoqu
  7

• Do you plan to release 320x512 resolution, 16 frames model?

  #5 opened 3 months ago by ebartrum
  8

• sh run.sh: IndexError: pop from empty list

  #26 opened 3 months ago by zhujiagang
  1

• error: Network is unreachable

  #25 opened 3 months ago by Ma0zii
  0

• Different aspect ratio

  #18 opened 3 months ago by sudonymously
  2

• Cannot find submodule dust3r.

  #24 opened 3 months ago by yuzhaoaaaa
  8

• Is it possible to support more camera movements?

  #15 opened 3 months ago by leeooo001
  2

• sparse view input

  #22 opened 3 months ago by sp77-1
  1

• AttributeError: 'VisionTransformer' object has no attribute 'input_patchnorm'

  #20 opened 3 months ago by O-O1024
  1

• Questions about the dataset.

  #17 opened 3 months ago by guiyuanyuanbao
  1

• Code release consulting

  #6 opened 3 months ago by Sarlren
  4

• Question about upcoming sparse-view novel view synthesis and iterative novel view synthesis

  #14 opened 3 months ago by tsuJohc
  2

• Questions about cases on dynamic video NVS

  #11 opened 3 months ago by xizaoqu
  8

• ERROE：pip install -r requirements.txt

  #7 opened 3 months ago by RyanSunn
  1

• Image Color changed

  #9 opened 3 months ago by LaFeuilleMorte
  1

• Work with other point cloud generation models and diffusion models?

  #13 opened 3 months ago by LearnedVector
  3

• It seems that only CUDA is supported at the moment. Are there any plans to support MPS?

  #12 opened 3 months ago by glovebx
  1

• Can We generate a single novel view given rotation angles?

  #8 opened 3 months ago by LaFeuilleMorte
  1

• Request: more optimized model which can run into low vram gpus

  #4 opened 3 months ago by BahzBeih
  1

• Using slow RoPE2D version?

  #3 opened 3 months ago by atonderski
  1

• Very exciting work! Some questions about this work.

  #2 opened 3 months ago by lebron12332
  2