facebookresearch/LaViLa

Issues

• About Ego4d dataset

  #37 opened 5 months ago by lwpyh
  2

• May be repeatedly loade the checkpoints.

  #32 opened 5 months ago by yyvhang
  1

• EGTEA reproduce

  #30 opened 9 months ago by jong980812
  1

• What is the source of WIT video dataset?

  #35 opened 6 months ago by rixejzvdl649
  3

• Checkpoint of the pre-trained dual-encoder.

  #36 opened 6 months ago by AlbertHuyb
  1

• val .pkl files for pre-training LAVILA Dual-Encoder

  #34 opened 6 months ago by AlbertHuyb
  2

• Run locally on multiple GPUs

  #29 opened 10 months ago by maximotus
  3

• Base narrator model

  #33 opened 8 months ago by sarisel
  0

• Currently unable to run the demo between Colab and Huggingface.

  #19 opened a year ago by kuangxiaoye
  3

• Cannot use Huggingface demo

  #31 opened 8 months ago by fgvfgfg564
  0

• if the model supports Chinese language prompts or narrations?

  #28 opened a year ago by ruifenggong
  1

• About the demo

  #17 opened a year ago by aa7784171
  1

• Extracting spatial feature maps from LaViLa

  #27 opened a year ago by vineetparikh
  0

• Preprocessing of Ego4D for pretraining

  #24 opened a year ago by chuyishang
  2

• The pretraining weights of TSF-B/L (visual only) on EPIC-KITCHENS-100 and EGTEA.

  #18 opened a year ago by daiguangzhao
  1

• Pretrained weight of HowTo100M

  #13 opened 2 years ago by HYUNJS
  3

• Training/fine-tuning the narrator

  #23 opened a year ago by tobyperrett
  2

• LaViLa as feature extractor

  #22 opened a year ago by deepsurbhi8
  3

• The pre-training weights of Dual-Encoder Baseline (with TSF-B/L)

  #20 opened a year ago by daiguangzhao
  2

• The pre-training weights of Dual-Encoder Baseline (with TSF-B/L)

  #21 opened a year ago by daiguangzhao
  1

• Question about preprocessing Ego4D

  #16 opened a year ago by hyojinie
  2

• Normalization values for CLIP models

  #15 opened a year ago by Jazzcharles
  1

• Narrator Training

  #12 opened 2 years ago by Flaick
  2

• Reproducing zero-shot eval results on EK100-MIR

  #11 opened 2 years ago by melongua
  1

• Thank you for contributing such excellent work!

  #8 opened 2 years ago by daiguangzhao
  1

• Train LAVILA (L) to perform action recognition on the EPIC-100 dataset?

  #9 opened 2 years ago by daiguangzhao
  1

• Training Time

  #3 opened 2 years ago by mmaaz60
  2

• Segmentation fault when launching demo_narrator [was: Keys remapping seems not to work]

  #4 opened 2 years ago by amessina71
  9

• Resized Version of EK100

  #7 opened 2 years ago by SJTUwxz
  3

• Training narrations for downloading

  #5 opened 2 years ago by melongua
  1

• Add models/demo to Hugging Face Hub

  #1 opened 2 years ago by nateraw
  4

• Git clone failing in Colab

  #2 opened 2 years ago by nateraw
  1