Maswali yanayounganishwa na mwongozo · Kati Kiwango

Maswali ya Uchoraji ya Azimio-Robust ya LaMa

Jifunze jinsi LaMa inavyochanganya michanganyiko ya Fourier, upotevu wa hali ya juu, na vinyago vikubwa vya uchoraji wa picha na mipaka yake iliyoripotiwa.

Njia za mwongozo zinazohusianaLama Azimio Robust Uchoraji
Swali 1 ya 8

Je, ni sehemu gani ya usanifu wa saini inayoipa LaMa mtazamo wake wa kimataifa wa picha?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. Je, ni sehemu gani ya usanifu wa saini inayoipa LaMa mtazamo wake wa kimataifa wa picha?
    • Haraka Fourier Convolutions
    • Vizuizi vya kujiangalia kwa transfoma
    • Safu za LSTM za kawaida
    • Mitandao ya capsule
  2. Kwa nini LaMa inashughulikia vinyago vikubwa bora kuliko mifano mingi ya awali ya uchoraji?
    • Inafanya kazi kwenye picha ndogo tu
    • Sehemu yake ya kimataifa ya kupokea huiruhusu kuendelea na muundo kwenye mashimo makubwa
    • Hunakili saizi nasibu kutoka kingo
    • Inakataa kujaza masks juu ya ukubwa uliowekwa
  3. What resolution behavior did the LaMa paper report?
    • It only works at one exact training resolution
    • It works only below the training resolution
    • It generalized to some resolutions higher than those used during training in the evaluated settings
    • It accepts no image input
  4. Ni aina gani ya maandishi ambayo LaMa hushughulikia vizuri shukrani kwa usindikaji wake wa kikoa cha mzunguko?
    • Kelele safi ya nasibu
    • Rangi moja tu thabiti
    • Nakala herufi pekee
    • Miundo inayojirudia, mara kwa mara kama vile matofali au vigae
  5. Compared with large diffusion-based inpainting systems, how is LaMa commonly positioned?
    • As a comparatively efficient convolutional inpainting method
    • As a much slower model that requires a data center
    • As a video-only model
    • As a tool that never needs an image
  6. Ni hasara gani husaidia LaMa kutoa vijazo vinavyoonekana kuwa halisi badala ya wastani wa ukungu?
    • Upotezaji rahisi wa pikseli wa maana-mraba pekee
    • Hasara ya adui pamoja na hasara ya utambuzi
    • Hakuna utendakazi wa upotezaji hata kidogo
    • Hasara ambayo hupima ukubwa wa faili pekee
  7. What additional loss component does the LaMa paper emphasize for large-mask inpainting?
    • Hasara ya utambuzi wa usemi
    • A bitrate-only loss
    • A perceptual loss with a high receptive field
    • A loss that ignores image structure
  8. Who introduced LaMa in the 2021 paper?
    • Jonathan Frankle and Michael Carbin
    • Cees Taal and colleagues
    • Timothy Dozat and Christopher Manning
    • Roman Suvorov and colleagues