Nhungamiro yehunyanzvi

Dilated uye Atrous Convolutions

Dilated convolutions (inonziwo atrous convolutions) isa maburi pakati pezviremu zvemasefa kuitira kuti kernel ivhare nzvimbo yakakura pasina kuwedzera paramita.

Teerera kugwaro iriAI-yakarondedzerwa · ~2 min

Tanga kuverengaTora mubvunzo
Pa peji ino2 min verenga
  1. Pfupiso
  2. Kudzika Kwakadzika
  3. Kuitwa muNyika yeChokwadi
  4. Ramangwana reDilated uye Atrous Convolutions
  5. Kuzvishandisa
  6. Ramba Uchiongorora
  7. Mibvunzo inowanzo bvunzwa

Pfupiso

Ivo vanoita kuti manettiweki aone zvakakura mamiriro, akakosha kune zvikamu uye odhiyo, uku ichichengeta kugadzirisa.

Kudzika Kwakadzika

Yakajairika convolution kernel inobata mapixels ari padyo. A dilated convolution inoparadzira huremu hwakafanana hwekernel nechiyero chekuwedzera, kusvetuka pixels pakati, saka 3x3 kernel ine dilation 2 inotenderera 5x5 dunhu ichiri kushandisa 9 huremu chete. Izvi zvinokurisa nzvimbo yekugamuchira zvakanyanya kana ukarongedza maseru nemitengo inowedzera, zvichiita kuti inetiweki iunganidze hukuru-hukuru pasina kubatana kana nhanho izvo zvinoderedza chimiro chemepu. Izwi rokuti atrous rinobva kuFrench trous, zvichireva kuti ane maburi. Izvi zvakakosha mumabasa akaomesesa ekufanotaura senge semantic segmentation, kwaunoda ese maonero akakura uye pixel-chaiyo kubuda, uye muWaveNet yekuenzanisira kureba kwekuteerera.

Kuitwa muNyika yeChokwadi

DeepLab inoshandisa atrous convolutions uye ASPP yemamiriro-of-the-art semantic segmentation yezviono zvemumugwagwa.

WaveNet stacks yakawedzera causal convolutions kuti ibudise echokwadi yakasvibira odhiyo uye kutaura

Segmentation yemifananidzo yezvokurapa, senge bundu kana miganhu yenhengo, uko kwakatambanuka pamwe neruzivo rwakanaka zvese zvine basa

Real-time parsing yekuzvifambisa-kutyaira iyo inoda minda yakakura yekugamuchira pasina kurasikirwa nekugadziriswa

Ramangwana reDilated uye Atrous Convolutions

Dilated convolutions inogara pakati pesemantic uye panoptic segmentation, kufungidzira kwekurapa, uye chizvarwa chekuteerera. Iwo ari kuwedzera kusanganiswa netarisiro, uko dilation inopa yakachipa yemarefu-refu minda yekugamuchira iyo inopindirana iwe pachako. Tsvagiridzo inoenderera mberi pane inoshanduka uye inodzidzika dilation mitengo uye pakudzivirira gridding zvigadzirwa. Vatarisire mumhando dzakareba-dzinotevedzana uye chaiyo-nguva yechiitiko kunzwisisa kune anozvitonga masisitimu.

Kuzvishandisa

Nei zvichikosha

  • Mutengo uye bhajeti. Zvisarudzo zvezvivakwa zvinotyaira kuita uye mutengo wekushandisa kwemakore.
  • Sarudzo dzakajeka. Dzidzo yehunyanzvi inobatsira zvikwata kusarudza murwi wakakodzera, kwete iwo mutsva chete.
  • Kudzora kwemhando yepamusoro. Sarudzo dzeinjiniya dziri nani dzinoderedza zviitiko zvekuvimbika mukugadzira.

Chenjera

  • Kugadzirisa imwe bhenji kunogona kuvanza yakafara system kushaya simba.
  • Infrastructure uye mari yekugadzirisa inowanzotarisirwa pasi.
  • Chengetedzo uye kucherechedzwa mapundu anogona kukura sezvo masisitimu anowedzera kuoma.

Matanho anotevera

  1. Tsanangura latency, mhando, uye mutengo zvinangwa usati waitwa.
  2. Benchmark pasi pechokwadi mutoro uye data mamiriro.
  3. Chishandiso chekutarisa zvikanganiso, kudonha, uye mushandisi maitiro.
  4. Gadzirira nzira dzekudzosera kumashure uye dzezviitiko usati wawedzera.

Ramba Uchiongorora

Maitiro eAI

Shandisa kuongorora nemazvo kana uchienzanisa sarudzo dzehunyanzvi.Verenga Guide

Kusimbisa Kudzidza

Enda zvakadzika mumatekiniki ekudzidzisa maitiro.Verenga Guide

Tsamba yemahara

Tora iyo yemazuva ese AI pfupiso

Nhatu dzakasimbiswa dzeAI nyaya mangwanani ega ega pakati pevhiki, dzakanyorwa neChirungu chakajeka. Yemahara zvachose, hapana ads.

Imwe email zuva rega rega revhiki. Bvisa kunyoresa kamwe chete. Hatimbotengese kana kugovera kero yako.

Zviedze iwe pachako

Take the Dilated and Atrous Convolutions quiz

Mhinduro yepakarepo pamhinduro yega yega, uye chitupa chinogovaniswa chine ID inobatika kana wapedza kosi.

Tanga mibvunzo

Tsigira dzidzo yemahara yeAI. AI Understanding ndeye 501(c)(3) isingabatsiri - hapana ads, hapana paywall, zvachose. Ita chipo

Mibvunzo inowanzo bvunzwa

Chii chinonzi Dilated uye Atrous Convolutions?

Dilated convolutions (inonziwo atrous convolutions) isa maburi pakati pezviremu zvemasefa kuitira kuti kernel ivhare nzvimbo yakakura pasina kuwedzera paramita. Ivo vanoita kuti network ione yakafara mamiriro, akakosha kune segmentation uye odhiyo, uku ichichengeta kugadzirisa.

Ko chiyero chekuwedzera chakakura kupfuura 1 chinoita sei kune convolution kernel?

Dilation inoisa huremu huripo nepakati, ichikudza nzvimbo yekugamuchira pasina kuwedzera huwandu hwema paramita.

Iyo 3x3 kernel ine chiyero chekuwedzera che 2 zvinobudirira inotora nzvimbo yakakura sei?

Ne dilation 2 pane imwe nzvimbo yakasvetuka pakati pehuremu hwega hwega, saka kernel inotenderera nzvimbo 5x5 ichiri kushandisa huremu 9 chete.

Sei dilated convolutions inonyanya kubatsira kune semantic segmentation?

Segmentation inoda yakakura mamiriro uye pixel-level ruzivo kamwechete; dilation inokura mamiriro pasina kudzikisira izvo kubatanidza kana kufamba kwaizokonzera.

Izwi rokuti atrous rinobva kupi?

Atrous inobva kuFrench trous, ine maburi, inotsanangura maburi akaiswa pakati pehuremu hwekernel.

MuWaveNet, nei akaturikidzana akadhindwa convolutions akakosha?

Kuwedzera kuwedzera kukwirisa mareti rega WaveNet modhi inotsamira pazviuru zvemasampureti eodhiyo asina hombe kernels.

Ramba uchidzidza

Mazano anoenderana

Mamwe madhairekitori akasarudzirwa nyaya iyi

Wana chitupa cheAI chemaharaKosi pfupi yemahara nebvunzo. Hapana mutengo, hapana kadhi. Inoda kusvika maawa 2.5.Tanga mahara →